THE HYDROCHEMICAL CHARACTERISTICS OF

GROUNDWATER IN THE INCOMATI ESTUARY

Shaheeda Adonis

B-Tech Analytical Chemistry(Pentech)

Submitted in partial fulfillment of the requirements for the degree of

MASTER OF SCIENCE

In the

Department of Chemistry

Faculty of Science

University of the Western Cape

September 2007

i

Acknowledgements

This work would not have been possible without the help and support of the

following people, which I would like to thank and acknowledge for their

contributions:

The CSIR for their financial support and allowing me time away from the

office to complete my work.

Ms. Christine Colvin who served as my internal supervisor. Thank you for

your support and for always having a open door and willing to listen and give

advise.

The analysts in the Marine Laboratory of the CSIR for competent analysis of

nutrients at lower detection limits.

My supervisors at the university, Prof. David Key and Prof. Charles Okujeni,

for their encouragement and guidance and critical analyses of my work.

To my family and friends for their support and encouragement.

Finally, I would like to thank my husband, Jason, for his patience, support and

understanding throughout my studies.

ii

Abstract

The Hydrochemical Characteristics of Groundwater in the

Incomati Estuary

The focus of this work was to monitor and evaluate the hydrochemical characteristics

of the groundwater in the Incomati Estuary for a period of one year. The aims of this

work were to:

To evaluate the groundwater chemistry data for any spatial and temporal

variations and

To evaluate the suitability of the groundwater for drinking and irrigation

purposes.

The methodology was based on chemical analysis of water samples taken from

boreholes located in a defined region of the Incomati Estuary sampled at different

depths at different time periods. From the statistical analysis of the chemistry data

(wide ranges and high standard deviations and medians), it is evident that there are

substantial differences in the quality/composition of the groundwater within the study

area.

Some degree of spatial variation was observed. Interpretation of Piper diagrams based

on results obtained show that it is evident that the boreholes located in close

proximity to the coast have a Na+/Cl- major ion composition. The high sodium and

chloride concentrations obtained in the boreholes close to the coast serve as an

iii

indication that the groundwater may be in contact with water of marine origin and

that there is a possibility of saltwater intrusion into the aquifer at locations close to

the coast.

Molar ratio plots of Cl and Na, however, were also used to investigate whether

salinization of groundwater occur close to the coast. These diagrams illustrate that the

boreholes close the coast have much higher Na/Cl molar ratios than the boreholes

located further inland. This therefore substantiates the fact that there is intrusion of

saltwater into the aquifer in the region close to the coast. A depth variation of TDS

was also observed. In all sampling periods a sudden increase in TDS was observed at

a depth of 43m. This increase can be attributed to a freshwater/saltwater interface at a

depth between 35m and 43m.

The boreholes located more inland, however, have a major ion composition

characteristic of freshly recharged groundwater (Ca2+/Mg2+/HCO3-). It can be

concluded that the boreholes located in the area of recharge(inland) are fresher than

those located in the area of discharge(at the coast). From isotopic analysis a trend of

TDS increase with isotopic enrichment was observed for certain samples, thus

indicating mixing of fresh groundwater with seawater. No temporal / seasonal

variations of the quality of the groundwater were observed over the study period.

From this study it is evident that the quality of the groundwater close to the coast is

poor due to high solute concentrations. Better quality groundwater is located more

inland. In addition, the deeper boreholes close to the coast have poorer quality water

iv

than those that are more shallower. Although there is poor quality groundwater in

certain areas of the Incomati Estuary, the groundwater does not pose an immediate

and direct threat to human health. It could however, have serious implications if

proper precautions and/or treatments are not employed soon.

v

TABLE OF CONTENTS

Page

Acknowledgements

Abstract

Table of Contents

List of figures

List of tables

Chapter 1: Introduction

1.1 Introduction

1.2 Aims of the study

Chapter 2: Groundwater in coastal environments: a literature review

2.1 Coastal aquifer groundwater

2.2 Groundwater and coastal ecosystems

2.3 The role of groundwater in a mangrove estuary

2.3.1 Groundwater fluxes in a mangrove estuary

2.3.2 Groundwater/surface water interface in a mangrove estuary

2.3.3 Salinity of groundwater in mangrove estuary

2.3.4 Groundwater as a nutrient source in a mangrove estuary

2.3.5 Effect of tidal hydrodynamics on groundwater in a mangrove estuary

2.4 The coastal ecosystem of Maputo Bay

2.5 Geology

2.6 Mangroves and Maputo Bay ecosystem

vi

Chapter 3 : Methodology

3.1 Field Sampling Methodology

3.2 Rationale for borehole sites

3.3 Water Analysis

3.3.1 pH and electrical conductivity

3.3.2 Alkalinity

3.3.3 Major cations

3.3.4 Major anions

3.3.5 Dissolved organic carbon (DOC)

3.3.6 Stable isotopes of hydrogen and oxygen

3.4 Accuracy check on chemical analyses

3.4.1 Ionic Balance

3.4.2 Piper Diagram

Chapter 4 : Analytical Results

4.1 Results

4.1.1 Major ions

a) Piper Diagrams

b) Bivariate Diagrams

c) Relationship between sodium and chloride

4.2 Isotope Analysis

4.3 Water quality assessment

Chapter 5 : Discussion and conclusion

References

List of abbreviations

vii

Appendix : Methods

1) pH Measurement

2) Electrical conductivity

3) Alkalinity

4) Major cations (K, Na, Ca and Mg)

5) Dissolved Organic Carbon(DOC)

6) Chloride and Sulphate

7) Orthophosphate

8) Ammonia

9) Nitrate

viii

LIST OF FIGURES Fig.1 Features affecting coastal aquifers

Fig.2 Map of Maputo Bay

Fig.3 Geological Map of the Study Area

Fig.4 Study area and model boundaries

Fig.5 Transect conceptual model

Fig.6 Map of sampling sites

Fig.7 Classification diagram for anion and cation facies

in terms of major-ion percentages

Fig.8 Piper diagram for March 2004 sampling run

Fig.9 Piper diagram for April 2004 sampling run.

Fig.10 Piper diagram for August 2004 sampling run.

Fig.11 Rainfall

Fig.12 Cl/HCO3 ratio vs. Cl for March 2004.

Fig.13 Cl/HCO3 vs. Cl for April 2004.

Fig 14 Cl/HCO3 vs. Cl for August 2004.

Fig.15 Na/Cl molar ratios for March & April 2004

Fig.16 Na/Cl molar ratios for August 2004

Fig.17 TDS for drilled samples

Fig.18 TDS vs sample location

Fig.19 Cl(mg/L) vs. Oxygen-18 for August 2004

ix

LIST OF TABLES Table 1 Results for March 2004

Table 2 Results for April 2004

Table 3 Results for August 2004

Table 4 Results – analytical statistics

Table 5 Sample Identification Table for March 2004

Table 6 Sample Identification Table for April 2004

Table 7 Sample Identification Table for August 2004

Table 8 Comparison of borehole water quality with the limits as set out by

DWAF & WHO for domestic purposes as well limits as set out by

DWAF for livestock watering and irrigation

Table 9 Calculated SAR values of samples compared to SAR limits of DWAF

1

Chapter 1 1.1 Introduction

Increasing urbanization is taking place along coastlines and estuaries and causing

increased use of groundwater that will have a large impact on the quality and quantity

of aquifer water (Campbell et al, 1992). To meet the ever-increasing water demand in

the world, groundwater is being extensively used to supplement the available surface

water. When one considers the water resources in areas bordering seas, coastal

aquifers become a very important source of fresh water (Ranjan, 2005). Groundwater

is increasingly being subjected to over-exploitation for agricultural, urban and

industrial uses, leading to the deterioration of groundwater in coastal areas through

saltwater intrusion. Additional problems such as the discharge of untreated or

inadequately treated wastewater, agricultural runoff from farms and discharge of

untreated sewage can all lead to the deterioration and contamination of groundwater

in coastal aquifer (Younsi, 2004).

Salinisation, however, is the most widespread form of groundwater contamination,

especially in coastal aquifers, and is represented by the increase of total dissolved

solids (TDS) and some specific chemical constituents such as Cl-, Na+, Mg2+, and

SO42- (Nadler et al.,1981; Magaritz and Luzier,1985; Dixon and Chiwell, 1992;

Morell et al.,1996; Sukhija et al.,1996 and Giménez and Morell, 1997). Potential

salinisation sources are diverse, including natural saline groundwater, seawater

intrusion, domestic, agricultural and industrial effluents. Among these sources,

seawater intrusion is the most common and widespread in coastal areas, and forces

the abandonment of water wells in many instances. Saltwater intrusion is natural

process, but it becomes an environmental problem when excessive pumping of fresh

water from an aquifer reduces the water pressure and intensifies the effect, drawing

salt water into new areas. Many of the coastal aquifers in the world already

experience saltwater intrusion caused by both natural as well as anthropogenic

processes. It therefore becomes necessary to understand the patterns of movement

and mixing between the freshwater and the saltwater as well as the possible factors

2

that can influence these processes (Ranjan, 2005). A case study was carried out at

Jeju volcanic island, Korea to identify the origin of saline groundwater in the eastern

part of the island. From the hydrogeochemical characteristics based on bivariate

diagrams of both major and minor ions is was evident that the changes in the

chemical composition of the groundwater in the study area are mainly controlled by

the salinisation process, followed by cation-exchange reactions (Kim et a, 2003).

Even though surface water is the main source of water in the urban areas,

groundwater is being heavily used for drinking water purposes (FAO Land and Water

Development Division, 2005). Most of the country’s population, however, (49%)

obtain drinking water from wells, whereas 29% is obtained from surface waters, 20%

from pipes and 1% from rain (Matimula, 2003).The effects and disruptions caused by

the civil war, which ended in 1992, are still being felt in Mozambique, especially in

the coastal areas. As more people returned to Mozambique after the civil war, the

demand for resources such as food and water increased. Only about 36% of the

population in Mozambique has access to safe drinking water whereas the other

communities have to make use of inadequate or contaminated water sources.

The densely populated peri-urban areas of Maputo however rely on the exploitation

of shallow groundwater. According to DAR (Department of Rural Water,

Mozambique), groundwater is estimated in certain areas to serve as many as 500

people in a radius of 500 meters. 35% of the population in rural areas and 30% in

small towns receive water pumped from boreholes. In the villages, where water

systems exist, the local authorities manage the water systems (Ibraimo, 1999).

In the central districts of the city of Maputo, where a public water supply network is

in place and water is piped in from outside the city, local groundwater still accounts

for a significant portion of the total consumption (Corbo et al, 1992). Groundwater

from the shallow aquifer below the city is mostly used as potable water and for

domestic purposes and to a lesser extent for the irrigation of family plots. In some

cases groundwater is also used for irrigation of agricultural plots or animal breeding.

3

A reported number of 600 wells was recorded for Maputo, although no reliable data is

available on the number of active wells of the present time(Corbo et al,1992). Most of

these wells are private wells, serving a single family or a small group of families.

Public wells are also scattered in the suburban districts of Maputo.

The pollution of surface and coastal waters is also an emerging problem in

Mozambique (United States Central Intelligence Agency, 2004), which can lead to an

increased demand for cleaner groundwater. Despite the fact that industrialization and

production levels in Mozambique are relatively low, the urban population, industry

and agro-industries are sources of pollution that can have a large impact on the

surrounding population. Since most of the industrial activities are located in heavily

populated areas, there are numerous examples of polluting activities, such as

petroleum refining industries situated in the major cities discharging untreated

effluent; industrial chemicals, fertilizers, pesticides, organic pollutants and paints

from the chemical industry, all of which discharge untreated effluent (Mendes et al,

2004).

Domestic water contamination in rural, urban and peri-urban areas is also widespread.

Adequate treatment of human wastes using drainage pipes and septic tanks occurs in

only 2% of families in Mozambique, of which 11% are in urban and less than 1% in

rural areas. About 65% of the families living in Mozambique do not have access to

any type of sanitary facility. They either have to defecate in open spaces or create pit

latrines. These are the types of practices that can significantly contribute to

groundwater contamination and water-bourne diseases in Mozambique (Matimula,

2003).

Although it has been long recognized that activities taking place within catchments

can and do have an impact on the coastal ecosystem, the linkage between human

activity in catchments such as farming, irrigation and water abstraction, ultimately the

impact on coastal resources is still unknown. The Tri-Partite Technical Committee,

which coordinates the regional water regulatory bodies in the Incomati Estuary,

4

recognised the need to integrate the inter-dependence of coastal function and

livelihoods with catchment management. The Incomati catchment or estuary is

extensively affected by the human activity, thus raising concerns that this may have a

detrimental but unquantified effect on the economically coastal resources. Based on

this, Incomati River-Maputo Bay system was then used as case study for this project

(Monteiro& Matthews, 2003).

For this reason the CSIR Catchment-2-Coast project was initiated (CSIR C2C

Brochure, 2003). The Catchment-2-Coast (C2C) project is a major research and

modelling project that is looking at the impacts of river catchment developments on

the sustainability of downstream coastal resources which provide important

livelihoods to urban and rural communities. Focused on the Incomati estuary and

Maputo Bay in Mozambique, southern Africa, the project aims to establish what

impact human activity in the catchment is having on the shrimp industry in Maputo.

Shrimp are an economically important resource in Maputo Bay and at one time were

responsible for the highest export earnings for Mozambique. Thirty-five per cent of

all export earnings in 1996 were derived from the shrimp industry and approximately

3000 artesnal and semi-industrial fishers derived a direct livelihood from this

resource (Hoguane, 1999).

The project took an interdisciplinary, systems approach to develop an understanding

of the linkages between the main domains of river catchments and their associated

coastal environments. Using a variety of models it seeks to translate human

development needs upstream into cost-based consequences at the coast.

This focus represents a departure from past practices. River management has

traditionally been focused on water supply and extraction for human needs in the

catchment. This means that economically important coastal resources, such as

Maputo Bay, are often affected in an unknown way by catchment projects that do not

explicitly internalise these costs. The Catchment-2-Coast project aims to remedy this

by designing, implementing and validating an integrated catchment-to-coast planning

5

and management support system, not only for the region under study, but for other

areas in southern Africa and beyond (Monteiro & Matthews, 2003).

The work described here forms part of the groundwater component of the Catchment-

2-Coast project. The objective of the groundwater component of the Catchment-2-

Coast project however is to define a relevant event scale of aquifer-based processes

that links catchment activities to the coastal zone and to quantify groundwater

physiochemical outputs. The focus of this work is, however, to monitor and evaluate

the hydrochemical characteristics of the groundwater in the Incomati Estuary over a

period of 1 year. The data will also be used to evaluate the suitability of the

groundwater for drinking and irrigation purposes.

1.2 Aims of the project

To monitor the hydrochemical characteristics of groundwater in the Incomati

Estuary over a period of one year to include both the wet and the dry seasons.

The data will be used to characterise the groundwater.

To evaluate the hydrochemical data for any spatial and temporal variation.

To evaluate the suitability of the groundwater for the purposes of drinking and

irrigation, with reference to recommended limits set by both DWAF and

WHO.

6

Chapter 2 Groundwater in coastal environments: a literature review

2.1 Coastal aquifer groundwater

A coastal aquifer is defined as a water bearing geological formation hydraulically

connected to the sea (Johannes, 1980). The movement of water in coastal aquifers is

generally towards the sea/ estuary. This is as a result of the positive hydraulic

gradient set up by the balance between recharge inland and discharge towards the

sea / estuary.

When dealing with the exploitation, restoration and management of fresh

groundwater in coastal aquifers, the key issue that arises is that of saltwater intrusion.

Coastal regions face many hydrological problems like flooding due to cyclones, wave

surge and drinking fresh water scarcity due to problem of saltwater intrusion. The

natural balance between freshwater and saltwater in coastal aquifers is disturbed by

groundwater withdrawals and other human activities that lower groundwater levels,

reduce fresh groundwater flow to coastal waters and ultimately causing saltwater to

intrude coastal aquifers (Barlow, 2003). Features which can affect coastal aquifers are

summarised in Fig 1

The extent of saltwater intrusion into a coastal aquifer depends on several factors,

including the rate of groundwater that is withdrawn compared to the total freshwater

recharge to the aquifer, the distance of the wells and drainage canals from the source

or sources of saltwater, the geologic structure and distribution of hydraulic properties

of the aquifer (Barlow, 2003). Saltwater can intrude into coastal freshwater aquifers

through several pathways, including lateral flow from the ocean, an upward intrusion

from deeper and more saline zones of the groundwater system as well as by

downward intrusion from coastal waters (Reilly & Goodman, 1987). Saltwater

intrusion reduces freshwater storage in coastal aquifers and can result in the

7

Fig.1 Features affecting coastal aquifers

abandonment of freshwater supply wells when concentrations of dissolved ions

exceed drinking-water standards. The degree of saltwater intrusion varies widely

among localities and hydrogeologic settings. In view of the fact that most of the

sediments of coastal aquifers are of marine origin, salt leaching can be seen as

another threat to the quality of groundwater in coastal aquifers. As a result of their

marine nature, the sediments have a high calcareous content and high alkalinity, of

which the calcareous content is mainly due to shells and calcareous algal fragments.

The dissolution of these components into groundwater leads to high salt content

(Rust, 1987).

Salinity is by far the most significant form of contamination in coastal aquifers. There

is also other forms of contamination, namely due to human influences that also pose a

serious threat to the quality of groundwater in coastal aquifers. Contamination due to

human influences can be attributed to:

Relative sea level rise Changes in hydrological regime

Land –subsidence: - anthropogenic changes in natural recharge: - natural precipiation evapotranspiration Coastal Zone Coastal Aquifers Rivers & estuaries

Coastal erosion seepage Shoreline retreat salinity surface water Tidal effect Effects

Human Activities Countermeasures

Ground water extaction -decrease groundwater extraction saline groundwater Land reclamation resources infiltration surface water

Lowering groundwater level -salt damage crops increase natu ral recharge -degradation ecosystem physical barriers -increase seepage inundation low -lying areas -saltwater intrusion

8

Sources designed to discharge substances into the earth, eg. Wastewater treatment

plants.

Sources designed to store and dispose of substances, eg. Graveyards, dumps and

pit lantrines

Sources designed to retain substances during transport, eg. pipelines.

Sources discharging substances as a consequence of planned activities, eg.

irrigation and fertilizer application (Wright, 1991).

2.2 Groundwater and coastal ecosystems

There are several ways in which groundwater interacts with and affects coastal

ecosystems. Groundwater discharge sustains the flow and aquatic habitats of coastal

streams during periods when surface runoff is low. Groundwater discharge also helps

to maintain water levels and water budgets of freshwater lakes, ponds and wetlands

(Kumar, 2003)

The role of groundwater in delivering contaminants to coastal water has become an

area of growing interest and concern. Although many types of chemical constituents

have contaminated coastal groundwater systems, much of the concern to date has

been focused on the discharge of excess pollutants into coastal ecosystems. Nutrient

contamination of coastal groundwater occurs as a consequence of activities such as

wastewater disposal from septic systems and agricultural and urban uses of fertilsers.

Excessive pollution can however lead to excessive production of algal biomass and

loss of important habitats such as mangroves (US National Research Council , 2000).

Groundwater reaches coastal environments either by direct discharge or as baseflow

in the streams and rivers that drain coastal areas. It is however becoming evident that

in some coastal environments, direct groundwater discharge can be a substantial

contributor of fresh water and dissolved constituents. It has been shown that

groundwater input into coastal waters accounts for as much as 65% of the total

freshwater inflows (Bokuniewicz, 1980).

9

2.3 The role of groundwater in a mangrove estuary

2.3.1 Groundwater fluxes in a mangrove estuary

The recharge and discharge of groundwater in estuaries are controlled by a

combination of periodic tidal and evapotranspiration fluctuations, irregular rainfall

and regional groundwater flow. Groundwater fluxes have been found to be important

factors affecting the wetland productivity through their influences on the

accumulation and removal of chloride, nutrients, toxins, hypersalinition, sediment

oxidation, pH and soil moisture content (Hughes et al, 1998). Based on the fact that

the fluxes of groundwater can also be seen to have an amplified influence on the

benthic communities and porewater environments, large scale phenomena such as

sediment erosion and deposition can be linked to offshore groundwater fluxes.

Burrows also play a significant role in the exchange between surface and

groundwaters in a variety of marine and freshwater environments. The flushing of

animal and crab burrows situated in the beds and banks of estuaries and ocean coasts,

provide a mechanism which can enhance water exchange and thus the transport of

materials such as salt, nutrients, oxygen and pollutants( Riedel et al, 1997)

The transport of salt is of vital importance in a mangrove system. Mangrove roots

absorb porewater from the sediments. This results in an increase in salt concentration

in the soil surrounding the mangrove roots. This soil then becomes highly

impermeable and greatly retards the diffusive transport of the salt away from the

roots. In mangrove swamps without burrows, the salt must diffuse from the roots to

the surface waters. However, with burrows, the distance over which diffusion occurs

is much shorter, resulting in a significant decrease in the diffusion time. The removal

of the burrow water then completes the salt removal process. As a high tide floods a

swamp, the water surface area is small due to vegetation, but has a significant slope.

The pressure difference that exists across the burrow opening, due to this slope,

causes a flow in the burrow that results in the flushing of the burrow water, thus

removing the salt. The burrow flushing process can provide a much faster mechanism

10

for salt-removal from the root to surface compared to diffusion from root to surface

(Ridd, 1996).

It has been observed that in the initial stages of the rising water at flood tide, a sheet

of water advances into a swamp from creeks that are often blocked by vegetation. The

first few millimeters of flooding water at a given point is not due to the water from

that sheet but is believed to come from the ground through numerous holes burrowed

by crabs. These crab burrows enhance the flow of groundwater. A swamp can also be

draining out through groundwater flow, as was observed in the Bashita – Minato

mangrove swamp in Japan (Mazda et al, 1990). After storms the tidal creek was

occasionally ponded by a sill at the mouth. It was found that when the creek was

ponded, the water level in the creek decreased by up to 10cm/day, which is about 15

times larger than the evaporation rate. It was then suggested that the creek was

draining out through groundwater flow. The flow also appeared to be limited to the

upper 90cm of the substrate, which corresponded to the maximum depths of crab

burrows (Mazda et al, 1990).

2.3.2 Groundwater/surface water interface in a mangrove estuary

It has long been known that groundwater and surface waters are intrinsically linked

systems. The areas around coastal environments, like estuaries, represent zones of

interaction and transition between groundwater and the estuary where dissolved

constituents such as pollutants, nutrients, etc. can be diluted, exchanged, transformed

and destroyed. To identify predominant processes affecting solute exchange across

transition zones it is critical to assess contaminant fluxes to the sediment/water

interface.

Groundwater / surface water interactions in estuarine environments are influenced by

a number of processes forming complex spatially and temporally variable systems.

Tidal activity often induces a fluctuating water table as well as an infiltration of

surface water into the coastal sediments, forming a “mixing” zone with groundwater

11

discharging from the adjacent aquifer. Even though there is no theoretical definition

for such a zone, the term groundwater / surface water interface is commonly used

(White, 1993). The transition of groundwater into the sediments and surface waters of

an estuary can represent a significant environmental change from the relatively stable

groundwater environment. Changes in surrounding sediment chemistry, biological

activity and groundwater flow paths can be expected to influence the movement and

fate of contaminants in the groundwater environment.

2.3.3 Salinity of groundwater in mangrove estuary

The salinity of groundwater can be defined as being the dissolved mineral

concentration in the water. The intrusion of seawater into an aquifer at its place of

discharge to the sea or the diffusion / leakage of saltwater through poorly permeable

deposits into an aquifer are but just a few of the natural ways through which the

salinisation of groundwater in coastal areas is brought about. The over-exploitation of

groundwater in coastal areas is clearly a crucial problem due to large population

growth and the fact that 70% of the world population occupies coastal regions .

Owing to the expansion of both aqua-culture and agricultural activities in coastal

areas like estuaries, the over-exploitation of the groundwater resources causes

groundwater levels to fall which is coupled with land subsidence.

Consequently the hydrostatic balance between the groundwater and seawater

interface can be broken, causing seawater intrusion into the aquifer and thereby

causing deterioration of the quality of the groundwater. In nature the groundwater–

saltwater interface seldom remains stationary. Large-scale recharging into the aquifer,

as well as withdrawals from the aquifer, result in the movement of the interface from

one position to another. The movement will be advancing or retreating depending on

whether the freshwater flow through the aquifer is decreased or increased. Changes in

groundwater recharge directly affect the changes in fresh groundwater resources.

Consequently, the salinisation of coastal aquifers will accelerate due to the reduction

of groundwater recharge. This can lead to a reduction of fresh groundwater resources.

12

The mangroves are well adapted to the high salinity of the surrounding water and

soils in an estuary. The salinity limit of mangroves in an estuary is considered to be

90 parts per thousand (Smith, 1988). An increase in salinity will significantly

decrease the leaf emergence, the growth rate as well as seedling mortality.

Groundwater, however, can play a very vital role in this regard. The discharge of

groundwater can be regarded as a very important flushing mechanism that can clean

the mangrove soil from salt and other contaminants left over after evapotranspiration.

The discharge of groundwater into an estuary has also been seen to dilute the salinity,

as was observed in False Bay, South Africa. The potential discharge of groundwater

seepage in False Bay was estimated to be between 0 and 120m3/day, diluting the

salinity from an average of 350/0 to 34.50/0 at the place of discharge (Giljam, 2002).

2.3.4 Groundwater as a nutrient source in a mangrove estuary

Groundwater may be an important source of nutrients to the coastal ocean and other

standing bodies of water (Valiela et al, 1990). Many dissolved nutrients and

constituents, both natural and anthropogenic, can be carried to the surface waters

through submarine groundwater discharge (SGD). Sewage, mining waste or other

soluble refuse percolating into an aquifer might eventually enter coastal waters via

groundwater discharge. Depending on the concentrations and flow paths of these

contaminated constituents, the ecology of the coastal waters could be impacted where

they can cause eutrophication and an increase in labile organic matter in the

ecosystem.

The concentration of many contaminants in groundwater is typically several times

higher than seawater even in pristine waters due to the fact that groundwater normally

has a different composition than that of receiving coastal waters. The inter-

relationship between estuaries and groundwater is complicated as tidal fluctuations

may cause pressure waves to propagate inland, causing groundwater levels and

hydraulic gradients to fluctuate (Serfes,1991). Furthermore, if a change in the water

level of an adjacent estuary results in an increase in the hydraulic gradient, there will

13

be an associated increase in groundwater discharge to the estuary and the concomitant

potential for increased nutrient loading (Staver & Brinsfield, 1996).

Due to the high organic content in mangrove sediments, high levels of organic

phosphorus and nitrogen can be present in mangrove sediment. The mangrove

sediments can remove phosphorus from sewage effluent but is limited by the

availability of exchange sites in the phosphorus rich sediments as well as by the

precipitation of ferrous hydroxide. Much of the nitrogen present occurs as organic

nitrogen, whereas the inorganic nitrogen occurs in the form of ammonium. The lack

of oxygen will further hinder the oxidation process. Nitrate is rapidly denitrified by

anaerobic bacteria, under reducing conditions to gaseous nitrogen. The interstitial

ammonium is highly mobile, making it very susceptible to leaching by heavy rains or

lost through drainage following tidal inundation, which leads to contamination of

groundwater. The uptake of nutrients by mangroves can also lead to the

immobilization of significant amounts of nitrogen and phosphorus in a mangrove

estuary. The nutrients absorbed by the mangroves would eventually be returned to the

ecosystem in the form of leaf litter and then most probably flushed away by the tide.

2.3.5 Effect of tidal hydrodynamics on groundwater in a mangrove estuary

Coastal aquifers are complex zones due to the combined influences of oceanic

oscillations and inland groundwater forces. These influences lead to complex

behaviour of subsurface contaminants discharging into the sea or estuary. Important,

unique factors of a coastal aquifer system include the influence of the density contrast

between the salt water and fresh groundwater and tidal fluctuations. Their net effects

lead to complicated hydrodynamic conditions in coastal aquifers, which in turn could

influence the local ecological environment (Ataie-Ashtian et al, 1999; Bear et al.,

1999). In coastal aquifers the groundwater level/water table fluctuates with time in

response to the water level fluctuations of the tidal water body (sea or estuary). This

natural phenomenon has the potential to provide a convenient, economic and reliable

14

way of identifying coastal hydrogeological conditions in large scale. This could be

one of the reasons why the study on the dynamic relation between seawater and

coastal groundwaters became such an active research area since the 1950’s. The

influence of sea tide on the mean groundwater levels became one of the aspects that

immediately relate to the estimation of submarine groundwater discharge (Moore,

1999).

As tidal water table fluctuations propagate inland their amplitude is attenuated and

phase-shifts can take place. A typical damping distance for tidal watertable

fluctuations in an unconfined aquifer is several hundred meters and in a confined

aquifer it can extend landwards by several thousands meters (Lanyon, 1982). These

tidal fluctuations were found to largely affect the groundwater flow and mass

transport in both a confined and an unconfined aquifer (Lie et al., 2000). Tidal

fluctuations can also be seen to affect the intrusion of seawater into an unconfined

aquifer since the tidal fluctuations of the sea level can influence the

groundwater/seawater interface. It was found that tidal activity can force seawater to

intrude inland and creates a thicker groundwater/seawater interface. In cases where an

aquifer extends to a river, lake or estuary the effect of tidal fluctuations on seawater

intrusion into the aquifer might be stronger due to the reduction of groundwater flux

attributed to the over-height in water table caused by tidal fluctuations (Ataie-

Ashtiani et al., 1999).

The fate of constituents like nutrients and heavy metals discharged into the mangrove

system are influenced by the frequency and duration of tidal inundation, as well as the

pattern of water movement and drainage in the mangrove system. In many cases the

frequency and duration of tidal inundation may have a greater influence on the

removal and immobilisation of nutrients and heavy metals in sediments than the

chemical and biological characteristics of sediments in a mangrove system. Mangrove

swamps help control the tidal hydrodynamics of many tropical estuaries. It was

determined that peak tidal velocities of more than 1m/s in the tidal creek of Coral

Creek, Australia were slowed down to less than 0.7m/s in the mangrove swamp

situated 50m from the creek. From statistical experiments it was suggested that the

15

destruction of vegetation or the recovery of mangrove land could reduce the peak

tidal current as well as the asymmetry of the tidal current. This would consequently

result in siltation of the channel (Wolanski, 1992). It is, however, important not to

neglect the effects of tidal fluctuations and hydrodynamics when one wants to

evaluate the quality of groundwater.

2.4 The coastal ecosystem of Maputo Bay

Maputo Bay receives inflow from the rivers Inkomati, Maputo and Umbelzi that flow

into the Indian ocean. The Maputo Bay ecosystem is characterised by the mangrove

forests. It is a highly eutrophic ecosystem that is very important to the coastline, as it

protects the coastline from erosion and it provides important artisanal fisheries to

local communities. The estuary also plays a crucial role in the economical and social

life of Maputo. The health and productivity of the Maputo Bay ecosystem is also

dependent on the quantity and quality of the freshwater inflow from the main in-

flowing source of freshwater – the Maputo River and also to some extent to the

quantity and quality of groundwater discharge.

The climate of Maputo can be characterised as tropical with summer rains. The mean

monthly temperature around Maputo is fairly stable with an annual average of about

23oC. The average maximum temperature varies from 25oC in July – August to 30oC

in January. The minimum temperature varies from 14oC in July – August to 20oC in

January.The average rainfall for Mozambique is 980mm and is about 800mm in

Maputo Bay (Hoguane,2002). There are high river flows for 3-4 months during the

summer season and low flows during the remaining part of the year in many of the

rivers. Droughts are also a common phenomena in Mozambique, especially in the

south. These can be related to the erratic distribution of the rainfall during the wet

season. The drought periods are often followed by heavy storms (Hoguane, 2002).

16

The capital of Mozambique, Maputo, lies on Maputo Bay. City residents rely

considerably on fishery resources, both for consumption and economic reasons.

Maputo Bay beaches also serve many residents and tourists as a leisure spot

throughout the year. Yet despite its beauty, there is growing evidence that the waters

inside the bay are polluted by untreated sewage coming from new developments in

the city that are not connected to the existing sewage and drainage facility and water

treatment plants.

Fig.2: Map of Maputo Bay

The population of Mozambique in 2003 was estimated by the United Nations at

18,863,000, which placed it as number 54 in population among the 193 nations of the

world. According to the UN, the annual population growth rate for 2000–2005 is

1.75%, with the projected population for the year 2015 at 22,537,000. It was

Maputo Bay

17

estimated by the Population Reference Bureau that 40% of the population of

Mozambique lived in urban areas in 2001 and the capital city, Maputo, had a

population of 2,867,000 in that same year.

Mozambique is one of the poorest and most aid-dependent countries in the world.

Much of Mozambique’s economy was devastated by almost three decades of internal

warfare during which millions of Mozambicans were displaced and many killed.

Characterized as a dual economy because of the large divide between the dynamic,

capital-intensive sector and under-performing traditional sector, Mozambique’s

economy is today relatively diversified. The economy of Mozambique heavily

depends on the exploitation of its fishery and agricultural resource base. All sectors

make an important contribution, although agriculture is the most important

contributor employing 80% of the population. Mozambique produces a surplus of

agricultural products. The main cash crops are sugar, copra, cashew nuts, tea, and

tobacco.

About 50 000 to 60 000 people in Mozambique are employed in the fisheries sector.

This makes a substantial contribution to the economy of the country. The fisheries

sector represents approximately 40% of the total export earnings. The most valuable

fishery resources of Mozambique are shallow and deep water shrimp, scad and

makerel (Hoguane,2002). The shrimp production in Mozambique has an average

annual catch of about 185 ton/year (estimated value 3 million US$ a−1). Fish catches

in Maputo Bay amount to 550 ton/year (estimated value 0.8 million US$ a−1) (Sengo,

2003).

2.5 Geology

The area of Maputo forms part of the southern sedimentary basin with eastward

dipping deposits. Almost all the deposits in the South Mozambique Sedimentary

18

basin are mainly of marine origin and have been formed during transgressions phases.

Within the study area the oldest layer is the Grudja Formation of Upper Cretaceous

age consisting mainly of clayey sandstones, marles and clays. On top of it the Lower

Tertairy Salamanga Formation was deposited. Around Maputo the lithology of this

Formation changes from calcareous sandstones in the West via argillaceous

sandstones in the central part to calcareous sandstones and sandy carstified limestones

in the East (IWACO/DNA, 1986). These formations were covered by dunes during

the Upper Pleistocene forming the present landscape.

From a hydrogeological point of view, the fine grained Grudja deposits can be

regarded as an impermeable base. The Salamanga deposits have moderate to high

transmissivities and are indicated as the second aquifer. The first aquifer is the dune

formation. In the coastal plain, the river valleys and its borders, the two aquifers are

separated by semi-permeable clayey layers thus creating semi-confined conditions in

the second aquifer.

.

19

Fig 3. Geological Map of the Study Area

20

2.6 Mangroves and Maputo Bay ecosystem

Mangroves are an important component of the interdial ecosystem of Maputo Bay, in

the form of nutrient source, nursery area, shore protection and habitat for other

species (Semesi, 1998). Due to the dense population around Maputo Bay, the

mangrove forests in these areas are under severe strain (Kalk, 1995). Mangroves are

being used by the local population for several different reasons. The trees are used for

fuelwood, either for domestic use or to sell. Mangrove wood is also used for the

construction of fishing boat and fishing poles (Hatton, 1995).

21

Chapter 3 Methodology This chapter discusses the methods used to collect and analyse groundwater samples

from the study site. The rationale for the selection of the borehole sites is also

described within this chapter.

3.1 Field sampling methodology

Groundwater samples were collected by technical staff of the CSIR from March 2003

to August 2004. Groundwater samples were collected following the sample protocol

for groundwater sampling set out by Weaver (1992). Most of the boreholes were

constructed for this study. Each borehole was pumped for a maximum of 30minutes

before samples were taken. This was done to ensure that fresh water was sampled

from the aquifer rather than stagnant water from the borehole. The electrical

conductivity of the groundwater was also measured during pumping and samples

only collected after 30 minutes and once a stable conductivity reading was obtained.

However, the same could not be done for the large diameter wells. Instead a 1L

Teflon bailer was used to do the sampling.

Water samples for the laboratory analysis were collected in 300ml screw cap plastic

bottles. For each sampling point, two plastic bottles were rinsed three times each with

the water to be sampled and filled to the brim, before sealing tightly to include as

little air as possible in the top of the bottle. The samples were then placed in cooler

boxes at a temperature of < 4oC. The samples were then placed in fridges and

transported back to the laboratories of the CSIR in Stellenbosch for analysis. Samples

for δ18 O (oxygen-18) isotopes analyses were collected in glass bottles to minimize

potential evaporation of the sample which could alter isotope ratios. Samples was

then stored and transported to the CSIR laboratories in Pretoria for analysis. The pH

and electrical conductivity were measured using a portable meter (Radiometer)

adjusted for water temperature.

22

B`

Figure 4: Study area and model boundaries

23

Fig.5: Transect conceptual model (approximately 18 500m length by 1000m width)

NOT DRAWN TO SCALE

Confining clay lenses Incomati alluvium Possible positions of freshwater/sea interface

Explanation

Brown and red sand dunes, recharge zones Sandstone inland & Calcareous sandstone in coastal plain Grey Clay, aquifer basement

Island

Infulene river

Seasonal wetland

Sea

Region under tidal influence

Drainage ditch Coastal sand

dune ridges

Inland sand dune ridges

RM2

RM1

RM River mouth

B' B

24

Fig.6: Map of sampling sites

3.2 Rationale for borehole sites

Fig 4 above illustrates the main lithologies and the boundaries of the transect

(B-B’) modelled for this project using Modflow. The modelling for this project was

done by Freeternity Rusinga, a geologist that formed part of the Catchment-2-Coast

(C2C) project. The approach to this study was firstly to characterise the dynamic

hydrogeological flow regime, with particular reference to discharge to the estuarine

mangrove areas. The conceptual understanding of the flow regime, based on field

visits and existing data is outlined in the diagram on page 22 (Fig.5). The monitoring

boreholes used for this study were then installed and located along this transect. Two

multi-piezometer boreholes and four additional hand-augered piezometer boreholes

were installed in the tidal zone. The multi-piezometer boreholes were at different

IIssllaanndd CCrreeeekk

25

depths, namely 7m, 13m, 16m, 35m and 43m. The hydrochemical sampling for this

study was done at the newly installed boreholes and existing community wells within

the area of interest (Fig.6).

3.3 Water analysis

For this study 38 water samples were collected, of which 35 were groundwater

samples and 3 surface water samples(Seawater, Creek Surface and Creek Bed). All

the water analyses, except the isotope analyses, were conducted by the author using

the facilities of the Cape Water Program Analytical Services Laboratories of the

CSIR in Stellenbosch. The isotope analysis was done by CSIR laboratories in

Pretoria. The analytical methods are described briefly in this section, but details of the

analytical procedures and data reliability have been confined to Appendix B.

3.3.1 pH and electrical conductivity

Water sample pH measurements in the laboratory were carried out using a

Radiometer PHM82 standard pH meter and GK 2401C combined electrode, together

with a stirrer. The pH reading was recorded while stirring. A Radiometer conductivity

meter with platinum electrode cell and a thermometer was used for the electrical

conductivity measurements. The measurements were taken after adjustment for

temperature.

3.3.2 Alkalinity

The method of potentiometric titration to a preselected pH, as set out in Standard

Methods (1995), was used for the alkalinity measurements. A Radiometer PHM 82

pH meter with a combined pH electrode, a stirrer and a Radiometer Autoburette

ABU 80 in conjunction with Radiometer Autotitrator unit TTT 80 was used for the

titration of the water samples. A 20mL sample was transferred with a volumetric

pipette to a beaker and placed on the stirrer. The electrode and tip of the burette was

26

placed in the sample. The solution was titrated with the standard 0,010N HCI to the

endpoint pH of 4,5.

3.3.3 Major cations (K+, Na+, Ca2+, Mg2+)

Major cations in natural waters are usually in proportion to the total dissolved

salts of that water. In other words, by determining the electrical conductivity

of a sample, an appropriate dilution can be made so that the sample falls

within the range of the working standards, which is 1mg/L to 100mg/L for Na+

and Ca2+ and 0.3 mg/L to 30mg/L for K+ and Mg2+. The water samples were

analysed by atomic absorption spectrophotometry. A flame atomic absorption

spectrometer consisting essentially of the following components was used :

(i) A stable light source, emitting the sharp resonance line of the element to be

determined. (ii) A flame system into which the sample solution may be aspirated at a steady rate

(iii) A monochromator to isolate the resonance line and focus it upon a

photomultiplier.

(iv) A photomultiplier to detect the intensity of light energy falling upon it, which is

followed by facilities for simplification and readout.

(v) A chart recorder for recording the output.

The operating procedure for the analysis are confined in Appendix B: B1.4.

3.3.4 Major anions (Cl-, SO4 2-, NOX

-, PO4 3- NH4

+)

Sulphates and chlorides in natural waters are usually in proportion to the total

dissolved salts of the water. So by determining the conductivity of a sample, an

appropriate dilution can be used so that the sample falls within the range of the

working standards. For working range for sulphates and chlorides are 5mg/L to

250mg/L and 5mg/L to 500mg/L, respectively. The autoanalyser system used for

27

analysis consists of the following: recorder, colorimeter, mixing manifold with

appropriate coils, proportioning pump and an automatic sampler.

The analyses for nitrate, ammonia and phosphate were also performed using an

autoanalyser. Some of the samples that were below the detection limits for nitrate and

ammonia (0.5mg/L) of the freshwater laboratory of the CSIR was analysed by the

author in the marine laboratory of the CSIR in Stellenbosch , where the detection

limit for nitrate and ammonia was 0.01mg/L.

3.3.5 Dissolved organic carbon (DOC)

The water samples for the analyses of dissolved organic carbon (DOC) were first

filtered through a 0.45µm microporous membrane filter prior to analysis. The

analysis was performed by an autoanalyser using the persulphate – ultraviolet

oxidation method for total organic carbon, set out in Standard Methods (1995) as

Method 5310C.

3.3.6 Stable isotopes of hydrogen and oxygen

The analyses for the samples collected for oxygen-18 and deuterium analyses were

done by the CSIR Laboratories in Pretoria using a dual-inlet mass spectrometer.

Water samples had to be converted to appropriate gases for measurement by the mass

spectrometer.

3.4 Accuracy check on chemical analyses

3.4.1 Ionic Balance

The ionic balance was determined for the results of the major cations and anions and

the results are presented in the table of analysis in Chapter 4. The ionic balance was

calculated by:

Balance = 200)/())/()/()/(×

+−

LmeqALmeqCLmeqALmeqC

28

Where C (meq/L) is the total cation milli-equivalents per litre

A (meq/L) is the total anions milli-equivalents per litre

The ionic balance or charge balance error gives an indication of the accuracy of water

analysis data. A fundamental condition of electrolyte solutions, such as groundwater,

is that on a macro scale the sum of negative charges is equal to the sum of positive

charges, i.e. that a condition of electro-neutrality exists (Freeze & Cherry, 1979).

When significant deviation from equality occurs there must be either analytical errors

in the concentration determinations or ionic species present at significant

concentrations that were not included in the analysis. Normally a charge balance error

of < 10% is considered to be acceptable. However, it should be remembered that

large errors in individual ion analyses may balance each other out.

For all the water samples analysed during March 2004 to August 2004, the ionic

balance was < 5%. This is an indication that the data sets obtained from to March to

August 2004 were of good quality and all 36 samples will be used to determine the

hydrochemical character of groundwater in the Incomati Estuary.

3.4.2 Piper Diagram

A need developed to find a more convenient way to refer to water compositions by

identifiable categories and this is where the concept of hydrochemical facies was

developed. Hydrochemical facies are distinct zones that have cation and anion

concentrations describable within defined composition categories.

Piper Diagrams are one of the most useful ways of representing and comparing water

quality. In the Piper Diagram cations, expressed as percentages of total cations in

milliequivalents per liter, plots as a single point on the left triangle and anions,

29

similarly expressed as percentages of total anions, appear as a single point in the right

triangle. These two points are then projected into the central diamond-shaped are

parallel to the upper edges of the central area. This single point is thus uniquely

related to the total ionic distribution. The Piper Diagram also conveniently reveals

similarities and differences among groundwater samples. Those samples with similar qualities will

tend to plot together as groups (Todd & Mays, 2005).

Ca Cl

Cations Anions

Figure 7: Classification diagram for anion and cation facies

in terms of major-ion percentages. Water types are selected

according to the area in which they occur on the diagram segments

(Frreze & Cherry, 1979).

30

Chapter 4

Analytical Results

Table 1: Results for March 2004

Site ID: Drill 7m Drill 13m Drill 16m Drill 35m Drill 43m Manual 1 Piezo Creek Village Ridge Island Sample Date; 19-Mar 19-Mar 19-Mar 17-Mar 19-Mar 17-Mar 17-Mar 19-Mar 19-Mar 19-Mar 18-Mar Sample ID: A B C D E F G H I J K Potassium as K mg/L 23 22 20 21 329 300 365 105 6.5 7.3 308Sodium as Na mg/L 173 351 198 144 6958 8327 10409 2540 79 19 8038Calcium as Ca mg/L 567 119 420 95 650 357 446 120 45 18 339Magnesium as Mg mg/L 0.1 0.1 0.5 0.1 670 1056 1350 300 6.8 5.5 936Ammonia as N mg/L 1.3 5.0 0.8 0.8 5.8 2.9 1.0 0.1 0.5 0.1 0.1Sulphate as SO4 mg/L 6.3 74 18 14 2137 1613 2097 617 30 12 1895Chloride as Cl mg/L 118 630 123 250 12750 15385 18750 4542 135 30 14250Alkalinity as CaCO3 mg/L 1599 190 1252 206 112 679 807 88 87 45 459Nitrate plus nitrite as N mg/L 104 681 116 220 113 1333 1050 65 87 681 50Ortho phosphate as P mg/L <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 2.79 <0.1 <0.1 0.95 <0.1Dissolved Organic Carbon mg/L 26 368 24 11 24 5.7 3.3 3.9 23 1.3 <1Conductivity mS/m (25°C) 7500 300 580 172 3500 4050 5000 1400 76 27 3950pH (Lab) (25°C) 12.3 11.2 12.3 11.5 9.8 7.3 7.7 7.4 6.9 6.9 7.1Hardness as CaCO3 mg/L 1417 298 1052 237 4381 5240 6674 1535 140 68 4702SAR as mmol?L 2.0 8.8 2.7 4.1 45.8 50.1 55.4 28.2 2.9 1.0 51.0Total Dissolved Solids (mg/L) 48000 1920 3712 1100.8 22400 25920 32000 8960 486.4 172.8 25280 % Difference 3.07 4.45 4.57 1.02 1.84 1.33 1.11 0.81 4.34 0.44 0.20CATIONS meq/L 36.51 22.12 30.19 11.59 399.04 474.79 595.51 143.85 6.44 2.37 451.50ANIONS meq/L 35.42 23.11 28.87 11.47 406.37 481.10 588.94 142.70 6.17 2.38 450.58

31

Table 2: Results for April 2004

Site ID: Drill 7m Drill 13 m Drill 16m Drill 35m Drill 43m

Manual 1

Island Creek

Island. Creek at P Piezo

Well water

New M A

New M B

New M C Creek Seawater

Sample Date: April 04 April 04 April 04 April 04 April 04 April 04 April 04 April 04 April 04 April 04 April 04 April 04 April 04 April 04 April 04

Sample ID: A B C

D E F G H I J K L M N O Potassium as K mg/L 20 19 14 22 333 274 159 160 286 6.7 305 308 114 114 109 Sodium as Na mg/L 194 541 174 216 9314 8188 4552 4533 8000 86 9000 9141 3250 3125 3063 Calcium as Ca mg/L 527 115 375 25 663 340 182 169 318 43 341 330 125 123 115 Magnesium as Mg mg/L 0.40 35 0.40 10 1089 1042 542 545 947 8.0 1084 1073 406 399 393 Ammonia as N mg/L 1.6 <0.1 0.6 <0.1 2.7 3.1 <0.1 <0.1 <0.1 0.59 0.12 0.12 <0.1 <0.1 <0.1 Sulphate as SO4 mg/L 4.5 27 8.9 28 2552 1445 1125 1105 1797 35 2246 2227 828 781 785 Chloride as Cl mg/L 136 905 81 346 16429 14405 7714 7771 13625 145 16000 16429 5800 5650 5600 Alkalinity as CaCO3 mg/L 1512 254 1160 55 92 754 129 129 449 83 101 71 103 96 96 Nitrate plus nitrite as N mg/L <0.1 <0.1 <0.1 0.34 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.44 <0.1 <0.1 Ortho phosphate as P mg/L <0.1 0.10 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 Dissolved Organic Carbon mg/L 23.0 110 17 8.0 10 7.8 1.8 2.1 0.8 23 2.9 2.3 2.2 2.9 2.9 Conductivity mS/m (25°C) 680 350 530 152 4550 3950 2300 2300 3900 150 4300 4300 1800 1750 1750 pH (Lab) (25°C) 12.4 7.6 12.3 9.0 8.5 7.1 7.6 7.6 7.3 7.1 7.1 6.7 7.6 7.7 7.7 Saturation pH (pHs) (20°C) 5.8 7.2 6.0 8.5 7.0 6.3 7.3 7.4 6.6 8.1 7.2 7.4 7.6 7.6 7.7 Total Dissolved Solids (Calc) mg/L 4352 2240 3392 973 29120 25280 14720 14720 24960 960 27520 27520 11520 11200 11200 Hardness as CaCO3 mg/L 1318 432 938 104 6139 5140 2686 2666 4694 140 5314 5239 1984 1950 1906 SAR as mmol/L 2.3 11.3 2.5 9.2 51.7 49.7 38.2 38.2 50.8 3.2 53.7 55.0 31.7 30.6 30.5 % Difference 3.57 4.73 4.06 4.64 3.50 3.23 4.98 3.99 4.27 4.30 1.08 0.20 0.55 0.15 1.23 CATIONS meq/L 35.39 32.65 26.71 12.00 536.50 466.09 255.73 254.54 449.11 6.76 505.47 510.16 183.94 177.82 174.08 ANIONS meq/L 34.17 31.18 25.67 11.47 518.37 451.49 243.60 244.77 430.71 6.48 500.09 511.18 182.93 177.56 176.23

32

Table 3: Results for August 2004

Site ID: Drill 7 m Drill 16 m Drill 35 m Drill 43 m Manual 1 Creek Surf Creek Bed New Well Pit A Pit B Sample Date: 11-Aug 11-Aug 11-Aug 11-Aug 14-Aug 15-Aug 15-Aug 11-Aug 12-Aug 12-Aug Sample ID A B C D E F G H I J Potassium as K mg/L 15 17 22 138 271 269 256 68 306 312Sodium as Na mg/L 215 244 218 3846 7808 7077 6885 1377 8320 8360Calcium as Ca mg/L 483 357 21 432 375 292 288 69 333 347Magnesium as Mg mg/L 0.54 0.05 8.9 433 1002 866 840 119 1018 1023Ammonia as N mg/L 1.2 1.0 0.50 4.4 2.2 <0.1 0.32 <0.1 0.21 <0.1Sulphate as SO4 mg/L 21 28 2 1068 1528 1736 1852 269 2176 2141Chloride as Cl mg/L 250 250 383 7333 14265 12794 12353 2294 15000 14853Alkalinity as CaCO3 mg/L 1333 1050 65 87 681 116 50 220 113 104Nitrate plus nitrite as N mg/L <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.19 0.13 0.90Ortho phosphate as P mg/L <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.26 <0.1 <0.1Dissolved Organic Carbon mg/L 25 21 4.4 3.0 5.5 1.8 1.0 17 1.1 1.3Conductivity mS/m (25°C) 630 520 142 2110 3900 3510 3430 770 4100 4050pH (Lab) (25°C) 12.3 12.2 9.3 8.5 7.3 7.5 7.3 8.3 7.6 7.7Saturation pH (pHs) (20°C) 5.9 6.1 8.5 7.1 6.3 7.2 7.6 7.5 7.2 7.2Total Dissolved Solids (Calc) mg/L 4032 3328 909 13504 24960 22464 21952 4928 26240 25920Hardness as CaCO3 mg/L 1208 892 88 2862 5062 4293 4179 662 5024 5081SAR as mmol/L 2.7 3.6 10.1 31.3 47.8 47.0 46.3 23.3 51.1 51.0Kjeldahl Nitrogen as N mg/L - - - - 4.6 0.42 1.8 8.3 5.8 1.9 % Difference 0.56 1.16 2.72 1.10 0.01 0.29 0.40 0.17 0.12 1.60CATIONS meq/L 33.95 28.97 11.83 228.33 447.84 400.49 389.55 74.87 470.13 473.14ANIONS meq/L 34.14 28.64 12.16 230.84 447.78 399.34 388.00 74.74 470.68 465.68

33

Table 4: Results – analytical statistics

Range Mean Median Standard Deviation

Concentration limits for groundwater for domestic purposes (DWAF)

Seawater

Major Cations (ppm)

K+ 6.5 - 365ppm 147ppm 114 129

50

390

Na+ 19.0 - 10409ppm 3921ppm 3188 3662

200

10 500

Ca2+ 18.0 - 663ppm 261ppm 290 181

150

410

Mg2+ 0.05 - 1350ppm 465ppm 403 453

100

1350

NH4+ <0.1 - 5.8ppm 1.4ppm 1 1.6

Ns

Ns

Major Anions (ppm)

HCO3- 45.0 - 1599ppm 377ppm 115 454

Ns

142

NO3- <0.1 - 1331 ppm 245ppm 87 393

10

Ns

Cl- 30.0 - 18750ppm 6969ppm 5725 6583

200

19 000

SO42- 2.0 - 2552ppm 929ppm 799 887

400

2700

PO43- <0.1 - 2.79ppm 1.0ppm 1 1.2

Ns

Ns

Physical Parameters

Conductivity mS/m 27 - 7500 2213 1800 1874

150

5000

pH (Lab) 6.7 - 12.4 9.0 8 1.9

5.0 – 9.5

Ns

Total Dissolved Solids (Calc) mg/L 173 - 48000 14167 11 520 11994

1000

32 000

Ns= Not specifc

34

Table 5: Sample Identification Table for March 2004

Site ID Location Project ID Colour Code

Drill 7m Drilled Multipiezo A

Drill 13m Drilled Multipiezo B

Drill 16m Drilled Multipiezo C

Drill 35m Drilled Multipiezo D

Drill43m Drilled Multipiezo E

Manual 1 Coastal F

Piezo Coastal G

Creek Coastal H

Village Inland I

Ridge Inland J

Island Coastal K

35

Table 6: Sample Identification Table for April 2004

Site ID Location Project ID Colour Code

Drill 7m Drilled Multipiezo A

Drill 13m Drilled Multipiezo B

Drill 16m Drilled Multipiezo C

Drill 35m Drilled Multipiezo D

Drill43m Drilled Multipiezo E

Manual 1 Coastal F

Island Creek Coastal G

Island Creek at P Coastal H

Piezo Coastal I

Well Water Inland J

New MA Coastal K

New MB Coastal L

New MC Coastal M

Creek Coastal N

Seawater Coastal O

36

Table 7: Sample Identification Table for August 2004

Site ID Location Project ID Colour Code

Drill 7m Drilled Multipiezo A

Drill 16m Drilled Multipiezo B

Drill 35m Drilled Multipiezo C

Drill 43m Drilled Multipiezo D

Manual 1 Drilled Multipiezo E

Creek Surface Coastal F

Creek Bed Coastal G

New Well Coastal H

Pit A Inland I

Pit B Inland J

Table 4 gives a brief statistical summary of the data collected over the study period. In this table,

the data collected over the study period are being compared to the guidelines for groundwater set

by DWAF for domestic purposes and that of seawater. From this table very wide ranges and high

standard deviations are easily regonised for most of the parameters. Such wide ranges for solutes

concentrations suggest that multiple sources and/or complex hydrochemical process take place to

generate the chemical composition. The high nitrate concentration (mean=245 mg/L) in the study

area reflects a possible degradation of groundwaters due to anthropogenic contamination.

The samples collected from the newly drilled multi-piezometer boreholes in the tidal area meet

most of the requirements set by DWAF for drinking and irrigation purposes, except for

conductivity and pH. All the other boreholes fail to meet the requirements. The standard

deviations and medians are also very high for all the parameters analysed. This further illustrates

37

the vast difference in quality/composition of the groundwater within the study area differs. By

comparing the values of the mean, median and standard deviation for the different parameters to

the requirements set by DWAF and that of seawater, it is evident that seawater plays a very

significant role in the quality / composition of the groundwater in the study area.

4.1 Results

4.1.1 Major ions

a) Piper Diagrams

Fig.8, 9 &10 show the plots of the hydrochemistry of the water samples sampled during March

2004, April 2004 and August 2004 respectively on piper diagrams

Fig. 8 Piper diagram for March 2004 sampling run

H B C D E A K F G J I

Cl+NO3

SO4

T.Alk Na+K

Mg

Ca AnionsCations

March 2004 Piper Diagram

Saline Water

Freshwater

38

Fig. 9 Piper diagram for April 2004 sampling run

Fig.10 Piper diagram for August 2004 sampling run

N

B C D E A H G I

O F J

Cl+NO3

SO4

T.AlkNa+K

Mg

Ca AnionsCations

April 2004 Piper Diagram

Saline WaterFresh Water

GF

BCDAEHIJ

Cl+NO3

SO4

T.AlkNa+K

Mg

Ca AnionsCations

August 2004 Piper

Saline Water

Freshwater

39

All the boreholes sampled over the sampling period are illustrated in the above Piper Diagrams.

Of all the boreholes, only three were sampled in March, April and August, namely Manual 1,

Drill 7m, Drill 16, Drill 35m, Drill 43m and Creek. The Drill samples are all located at the same

borehole, but sampled at the different depths. The following trends and characterization of the

groundwaters in the study area were done by considering only these three boreholes / six different

samples in the Piper Diagrams:

• Most of the samples are illustrated as being Na-Cl type waters, except the drill 7m and

drill 16m being Ca-HCO3 type groundwaters.

• From the above Piper Diagrams it is observed that all or most of the samples plotting in

the saline region of the Piper Diagram, are the ones that are located in close proximity to

the coastline and at greater depths(drill 35m and 43m), whereas the Ca-HCO3 are at a

shallower depth (drill7m and 16m).

• When comparing the 3 Piper Diagrams one notices the difference in chemical

composition of the drill 35m sample. In Fig.8 the drill 35m plots in the middle of the

diamond shape within the Piper Diagram, whereas in Fig.9 and Fig.10, it plots in the

saline / Na-Cl region. The difference in chemical composition of Drill 35m in the March

Piper Plot compared to the April and August Piper Plots can be as a result of the rain

experienced during the Wet Season (November to April), causing the groundwater at a

depth of 35m to be influenced by the infiltration of the rain or due to a possible drop in the

water levels towards the end of the Wet Summer Season in April and during the mid Dry

Winter Season (See Fig.11).

• The fact that drill 35m plots in the saline region for both the April and August can

possibly also be attributed to the high abstraction that would normally be encountered

during the Dry Seasons (May to October).

40

• Besides for the difference in composition observed for drill 35m sample, no real seasonal

variation is observed. Some spatial variation is however observed where the boreholes

located a little further from the coast and at shallower depths have a composition

resembling “fresh” groundwater compared to the boreholes situated near the coastline and

at a greater depth.

Rainfall

0

2

4

6

8

10

12

14

16

18

20

22

24

JAN.O

4

FEB.04

MAR.04

APR.04

MAY.04

JUNE.04

JULY.04

AUG. 04

SEPT.04

OCT.04

NOV.04

DEC.04

Time (months)

Rai

nfal

l (cm

)

Rainfall

Fig.11 Rainfall

b) Bivariate Diagrams

The changes in the chemical composition of the groundwater can be caused by mixing with

saltwater as well as by chemical reactions, such as ion-exchange processes taking place within the

groundwater system. A hydrogeochemical study done in Korea by (Kim,2003) was used to

identify the origin of the saline groundwater in Korea. He made use of hydrogeochemical

characteristics based on bivariate diagrams of major and minor ions to show that changes in

chemical composition of groundwater are controlled by the salinization process followed by

cation exchange processes. The following bivariate diagrams, namely Fig.12 to Fig 14, illustrate

41

the Cl/HCO3 ratio vs. Cl relationship for the 3 sampling periods. These graphs allow one to

identify the boreholes with “fresher” groundwater and those influenced by the intrusion of saline

water.

Cl/HCO3 ratio vs Cl(mg/L) for March 2004

1

10

100

1000

10000

1.0 10.0 100.0 1000.0 10000.0 100000.0

Cl (mg/L)

Cl/H

CO

3 ra

tio

A B C D

E F G H

I J K

Fig 12: Cl/HCO3 ratio vs Cl(mg/L) for March 2004

Cl/HCO3 ratio vs. Cl(mg/L) for April 2004

1

10

100

1000

10000

100000

1.0 10.0 100.0 1000.0 10000.0 100000.0Cl (mg/L)

Cl/H

CO

3 ra

tio

F G A

B C D

E J O

H K L

M N I

Fig 13: Cl/HCO3 ratio vs Cl(mg/L) for April 2004

42

Cl/HCO3 ratio vs. Cl(mg/L) for August 2004

1

10

100

1000

10000

100000

1.0 10.0 100.0 1000.0 10000.0 100000.0

Cl (mg/L)

Cl/H

CO

3 ra

tio

A B

C D

E F

G H

I J

Fig 14: Cl/HCO3 ratio vs Cl(mg/L) for August 2004

From the graphs one sees that some boreholes have much higher ratios compared to the others.

The boreholes plotting in the lower ratios can be characterised as having “fresher” or more dilute

groundwater, whereas the boreholes plotting in the higher ratios as saline groundwater, that has

been influenced by the intrusion of seawater (Kim, 2003). These graphs further illustrates that the

boreholes in close proximity of the coastline and at greater depths are influenced by the intrusion

of saline water.

c) Relationship between sodium and chloride

As we investigate the quality and/or composition of groundwater in the study area that is

determined by saline water and/or water–rock interactions, the chemistry of the water can be

explained with respect to plots of dissolved species vs. chloride, due to the fact that chloride

behaves conservative as far as the salinization process is concerned. For this study the molar

43

ratios of sodium vs. chloride was plotted to determine or investigate the chemistry of the

groundwater in the study area.

Na/Cl molar ratios for March 2004

050

100150200250300350400450500

0 100 200 300 400 500 600Cl (mmol/L)

Na(

mm

ol/L

)

C D E

A F H

K B G

I J

Na/Cl molar ratios for April 2004

0.00

50.00

100.00

150.00

200.00

250.00

300.00

350.00

400.00

450.00

0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 450.00 500.00

Cl (mmol/L)

Na

(mm

ol/L

)

F I

A B

D E

N C

G H

J K

L M

O

Fig 15. Na/Cl molar ratios for March & April 2004

44

Na/Cl molar ratios for August 2004

0.00

50.00

100.00

150.00

200.00

250.00

300.00

350.00

400.00

0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 450.00Cl (mmol/l)

Na

(mm

ol/L

)

E F G

A B C

D H I

J

Fig.16 : Na/Cl molar ratios for August 2004

As shown in figures 15 and 16, the coastal boreholes and the drill 43m sample are the ones with

high Na/Cl molar ratios compare to the other boreholes at a distance further from the shore and at

lesser depths. The boreholes having the high Na/Cl molar ratios are also the ones with the higher

TDS values (Figs. 17 & 18).The high TDS value encountered in Sample H during the March

2004 sampling run can possibly be attributed to anthropogenic or point source contamination The

combination of the high Na/Cl molar ratios, combined with the high TDS values again serves as

evidence that seawater intrusion could be main cause of salinization of these boreholes. This trend

is observed in all three plots.

45

TDS for drill samples

0

10000

20000

30000

40000

50000

60000

Drill 7m Drill13m Drill 16m Drill35m Drill 43m

Site ID

TD

S (m

g/L

)

Mar-04

Apr-04

Aug-04

Fig.

17: TDS for drilled samples

TDS vs Sample location

0

10000

20000

30000

40000

50000

60000

Drill 7m Drill13m Drill 16m Drill35m Drill 43m Manual Creek Island

Site ID

TD

S (m

g/L

)

Mar-04

Apr-04

Aug-04

Fig.18: TDS vs sample location

4.2 Isotope Analysis

The naturally-occurring stable isotopes, δD and δ18O have become useful tools in groundwater

studies. The main hydrogeological use for environmental isotopes can be summarized as follows:

To identify the occurrence of mixing two or more water types

To provide signature to a particular water type

46

To provide residence time information about groundwater

Isotopes of oxygen and hydrogen are in some instances, ideal geochemical traces of underground

water because their concentrations are not subject to changes by interaction with the aquifer

material (UNESCO, 1973). Once the water is underground and removed form zones of

evaporation, the isotope ratios do not change, except by mixing. When precipitation infiltrates to

recharge the groundwater, mixing in the unsaturated zone controls the isotopic variations. The

water in the saturated zone then has a composition corresponding to the mean isotopic

composition of infiltration in the area. Analyses of δD and δ18O can be used to identify the

probable source of an underground water.

Cl - δ18O relationship for August 2004

-4.40

-3.90

-3.40

-2.90

-2.40

-1.90

-1.40

-0.90

-0.40

0.10

0.60

0 100 200 300 400 500

Cl (mmol)

O18

(0 / 00)

A B

C D

H J

G E

Fig.

19 Cl(mg/L) vs. Oxygen-18 for August 2004

The Cl - δ18O relationship reveals that three of the samples from the August 2004 sampling period

follows a trend of TDS increase with isotopic enrichment, indicating mixing of fresh groundwater

47

with seawater (Faye et al, 2004). One the samples, namely L (Drill43m), has a trend of high TDS

without isotopic enrichment. Other processes whose effects can be combined can be used to

explain this trend. Among them are the following:

(1) mixing process of different waters along flow paths, combined with evaporation,

(2) the isotopic contents of the local rainfall

(3) transpiration from vegetation cover is also known to abstract water without causing isotopic

fraction

(4) dissolution of the salt minerals could change the interstitial water chemistry, thereby

increasing the TDS in the transition zone (Gat, 1981).

4.3 Water quality assessment

The quality of most of the boreholes sampled in this study can be described as poor, based on

guidelines set out for drinking water purposes by the World Health Organisation (WHO) and the

Department of Water Affairs and Forestry (DWAF), even though a large number of the

population living in Maputo rely on groundwater as water supply. High solute concentrations

were obtained in most of the boreholes sampled over the sampling period. These high

concentrations limit the potential use of these water sources.

From the data in the table below, it is evident that the boreholes located further from the coast and

at shallower depths are of a better quality than those located near the coast, according to the

specification of DWAF and WHO as illustrated in Table 1 above. The boreholes near to the coast

and at greater depths have TDS and chloride concentration much higher than the boreholes

located further inland and at shallower depths. Due to the high TDS(>10 000mg/L) and

chloride(> 5000mg/L) values of these boreholes, the groundwater has a very salty taste, making it

almost impossible to use for drinking purposes.

48

Table 6 :Comparison of borehole water quality with the limits as set out by DWAF & WHO for

domestic purposes as well limits as set out by DWAF for livestock watering and irrigation (All

values are expressed as mg/L)

TDS SO42- Cl- Na+ Ca2+ Mg2+ Hardness

DWAF (drinking) 450 100 100 100 32 30 50

WHO (drinking) 1000 500 250 200 500

DWAF(Livestock) 100 1000 1500 200 1000 500

DWAF(Irrigation) <100 <70 0.2

Sample Location

Drill 7m (P7) 4192 10.6 168 194 526 0.35 1314

Drill 13m (P13) 2080 50.5 767 446 117 17.6 365

Drill 16m (P16) 3477 18.3 151 205 384 0.32 961

Drill 35m (P35) 994 15 326 193 47 6.3 143

Drill43m (P43) 21675 1919 12171 6706 582 731 4461

Manual (Coastal) 25387 1529 14685 8108 357 1033 5147

Piezo (Coastal) 28480 1947 16187 9204 382 1148 5684

Creek (Coastal) 10080 699 5096 2832 121 350 1742

Village (Inland) 486 30 135 79 45 6.8 140

Ridge (Inland) 172 12 30 19 18 5.5 68

Island (Coastal) 25280 1895 14250 8038 339 936 4702

Well water (Inland) 960 35 145 86 43 8.0 140

New Well (Coastal) 4928 269 2294 377 69 119 662

According to the requirements by the World Health Organisation (WHO) and the Department of

Water Affairs and Forestry(DWAF) for groundwater for the purpose of domestic use, most of the

objections to the groundwater quality are based on taste and aesthetic considerations, rather than

the risk of adverse health effects (WHO, 1994). Groundwater having high chloride, sodium and

TDS concentrations may indicate contact with water of marine origin (Bouwer 1978). Due to the

poor quality of the groundwater close to the coast, the people living in these areas are therefore

forced to travel long distance to obtain fresher groundwater from the boreholes that are located

more inland.

49

Most of the samples, with the exception of a few, also exceed the required or recommended limits

for livestock watering and irrigation purposes. Due to the high solute contents of most of the

groundwater in the study area, the greatest risk or threat in terms of irrigation is the risk of soil

degradation because of the high salinity of these waters. Sodicity hazard, for soil degradation, was

also evaluated by calculating the sodium absorption ratio (SAR) in mmol/L, using the formula

(Sposito, 1989) :

SAR = [Na+]

[Ca2+ + [Mg2+]

2

where [Na+] [Ca2+] and [Mg2+] are the concentrations of these cations in mmol/L.

The SAR values for the boreholes sampled over the study period are as follows:

The following is a sample calculation for two of the sampled boreholes:

SAR calculation for Drill 13m in March 2004

SAR = [15.3]

[6.0 + 8.3 X 10-3 ]

2

= 8.8 mmol /L

SAR calculation for Creek in April 2004

SAR = [135.8]

[6.1 + 33.2 ]

2

= 30.6 mmol / L

50

The recommended or allowable SAR limit according to guidelines set by DWAF for groundwater

is 3-4 mmol/L. See table below for the calculated values of the SAR ratios for the sampled

boreholes.

Table 7: Calculated SAR values of samples compared to SAR limits of DWAF

Sample ID March 2004 April 2004 August 2004 DWAF SAR Allowable

Limit 3-4mmol/L 3-4mmol/L 3-4mmol/L

Drill 7m 2.0 2.3 2.7 Drill 13m 8.8 11.3 Drill 16m 2.7 2.5 3.6 Drill 35m 4.1 9.2 10.1 Drill43m 45.8 51.7 31.3 Manual 50.1 49.7 47.8 Piezo 55.4 50.8 Creek 28.2 30.6 Village 2.9 Ridge 1.0 Island 51.0

Island Creek Piezo 38.2 Well Water 3.2

New MA 53.7 New MB 55.0 New MC 31.7

Creek Surface 47.0 Creek Bed 46.3 New Well 23.2

Pit A 51.1 Pit B 51.0

51

Chapter 5 : Discussion and Conclusions Although some of the findings in this study have already been discussed in chapter 4, the main

findings and conclusion will be summarised in this chapter and placed in the context of the

original study objectives. The original study objectives, stated in chapter1, were to evaluate the

hydrochemical characteristics of the groundwater in the Incomati Estuary over a period of one

year to include both the wet and the dry seasons. The data obtained over this period will be used

to characterise the groundwater. This second objective was to establish whether any spatial or

temporal variations in groundwater quality occurred. The third objective was to evaluate the

suitability of the groundwater in the area for drinking water and irrigation purposes and to also

establish if the groundwater could have an effect on the estuary. A total of 36 samples were taken

and analysed by the author over the study period. High solute concentrations were obtained in

most of the boreholes over the sampling period.

Major ions

Piper Diagrams

The Piper Diagrams were used to evaluate differences in major ion chemistry in groundwater

flow systems. From the Piper Diagram (fig9, 10 & 11) it is clear that different water types are

present in the groundwater system. As shown in all three Piper Diagrams, most of the samples

plot in the chloride and sodium regions of the two triangle plots respectively. A difference in

chemical composition was however observed for one of the samples namely Drill35m. In the

March 2004 Piper Diagram, Drill 35m plots midway of the saline region and the freshwater

region in the diamond shape of the Piper Diagram. As discussed earlier, this could be attributed to

the rain experienced during the Wet Season, causing the groundwater to be diluted due to

infiltration.

Besides for the drill 35m sample, no real seasonal variation was observed over the study period.

Some degree of spatial variation was however observed. As shown by the diagrams, the boreholes

located a little further from the coast appear to plot in the freshwater region and those closer to

the coast in the saline region. This further indicates that groundwater in close proximity of the

52

coast are influenced by the intrusion of saltwater. When considering the Piper Diagrams (Figs 9,

10 &11) the following conclusions can be drawn:

In view of the fact that sodium and chloride are the dominant ions in the boreholes located

close to the coast and at greater depth, it therefore serves as evidence that there is an

intrusion of saltwater into the aquifer at locations close to the coast and at a certain depth.

Bouwer (1978) also found that boreholes containing high chloride and sodium

concentrations can serve as an indication that the groundwater is in contact with water of

marine origin.

The boreholes appearing in the “mixed” region have a Na+/Ca2+/Mg2+-Cl- major ion

composition. A major ion composition of this nature is characteristic of stagnant

groundwater zones or of groundwater at some distance along a flow path.

The boreholes plotting in the freshwater region have a Ca2+/Mg2+-HCO3-major ion

composition. This is characteristic of freshly recharged groundwater that has equilibrated

with CO2 and soluble carbonate minerals under open system conditions in the soil zone.

From all three Piper Diagrams it is evident that the boreholes located in the area of

recharge (boreholes located inland) are fresher than the ones located in the area of

discharge (boreholes near the coast).

Cl-/HCO3- ratio vs Cl- and molar ratios of Cl- vs Na+

Graphs representing the relationship between Cl-/HCO3- ratio vs Cl- and bivariate diagrams of Cl-

vs Na+ and Cl- vs SO42-

(Figs.13 to 17) were plotted to distinguise mixing mechanisms with saline

water from chemical reactions in the groundwater system.

From the Cl-/HCO3- ratio vs Cl- graphs ( Figs. 13 to 15) it is evident that the boreholes at

shallower depth and inland have a much lower ratio compared to the boreholes at the coast and at

greater depths. This further supports the evidence of saltwater intrusion into the aquifer at

boreholes close to the coast and at a depth greater than 35m (Kim, 2003).

53

The molar ratio plots of Cl- vs Na+ illustrates a strong correlation between chloride and sodium.

This strong correlation indicates that such ions are derived from the same source of saline waters.

However, the boreholes with the higher molar ratios are representative of the more saline

groundwater and the boreholes plotting in the lower molar ratios of the more “fresher”

groundwater. These diagrams therefore further substantiate the fact that there is salinisation of

groundwater in the region close to the coast (Faye, 2004).

Even though no seasonal trends were observed in total dissolved solids (TDS) from the boreholes

sampled over the sampling period, a spatial variation (Fig.19) as well as a vertical variation was

observed in the TDS (Fig.18) values of the samples collected over the sampling period. The

graphs further supports the fact that the boreholes located near the coast and those at a greater

depth (> 35m) have much higher TDS values than the other samples. This trend in TDS values

can be attributed to the possible intrusion of saltwater or as a result of evaporation and

condensation in these areas. The reason, however, for the sudden increase in TDS at a depth of

43m serves as indication that the freshwater/ saltwater interface exists at a certain depth between

35m and 43m.

Groundwater quality

From the data that was collected over the monitoring period it is evident that the main processes

controlling the hydrochemistry of the groundwater in the Incomati Estuary are mixing and

dissolution reactions, respectively.

Effect on the Estuary

What effects do groundwater quality have on the Incomati estuary? The Incomati estuary

ecosystem depends upon its mangrove forests. The mangroves trees are believed to be very well

adapted to the high and variable salinities of its surrounding water.

Although it is believed that estuaries receive their primary freshwater input from fluvial

discharge, there is good evidence that direct groundwater discharge maybe responsible for up to

10% of the total freshwater input to estuaries and the ocean (Garrels & Mackenzie, 1979). From

the results obtained in this study it is clearly shown that the nutrient loading through groundwater

54

discharge will not be greatly impact the ecosystem of the Incomati Estuary, since the highest

obtained values were 4.1mg/L for PO43- and NO3

- and 5.8mg/L for NH3.

The intertidal and subtidal sediments in estuaries consist of a biogeochemical dynamic. A range

of chemical transformation and exchanges between the sedimentary material and interstitial

groundwater are thus expected as water passes the sediments and the adjacent beachface (Ullman,

2003). It can thus be concluded that the groundwater that is discharged will therefore have a

different composition to the groundwater confined in the aquifer. To what extent the mangroves

depend on the discharge of the groundwater is unknown and serves as grounds for further

research.

Groundwater use for drinking and irrigation purposes

What does the quality of the groundwater mean for the people living in and around the Incomati

estuary? From the results it is evident that the boreholes located close to the coast and at greater

depths, due to their high solute concentrations, are not suitable for domestic purposes, as they do

not comply to the limits set by the WHO and DWAF. The inland boreholes and the groundwater

sampled from shallower depth are of a much better quality. No microbiological analyses were

performed on the samples, but a H2S test was used in the field and if gave us an indication that

some form of bacteria are present. The source for this could the use of on-site sanitation systems,

mostly the pit latrines that are extensively used in the rural areas (Samaja, 1993).

As urban and industrial development continues to expand around the coastal areas of

Mozambique, especially Maputo, so to will the demand for fresh water. A proper desalination

treatment system should be implemented as treatment for the more saline groundwaters around

the coastal areas. Above and beyond the high salinity obtained in certain boreholes, the

groundwater in the area does not appear to pose an immediate and direct threat to human health.

A monitoring program should be employed to monitor for future groundwater pollution.

Due to the high solute concentrations, the boreholes close to the coast are also not suitable for

irrigation and livestock watering purposes, as they do not comply with the limits set out by

DWAF. The inland boreholes, however all fall within the limits set by DWAF for irrigation and

55

livestock watering purposes. The SAR values obtained for the all samples also fall outside the

limit of 10 set by DWAF. If these waters are not properly treated by making use of proper

desalination treatment plant, it can lead to the degradation of the soil in the area and this can have

detrimental effects on the people living in these areas as they depend on the selling of their crops

for survival.

I

REFERENCES

Alaie-Ashtian, B., Volker, R.E. & Lockington, D.A. Tidal effects on seawater intrusion in

unconfined aquifers, Journal of Hydrology, Vol.216, pp.17-33.

Bokuniewicz, H (1980) Groundwater seepage into Great South Bay, New York, Shelf

Science, vol.10, pp. 437-444

Bouwer, H. (1978) Groundwater Hydrology, McGraw –Hill, New York,

Buddemeiser, R.W. (1996) Groundwater discharge in the coastal zone: Proceedings of an

International Symposium, LOICZ Reports and Studies no.8,LOICZ, Texel, The

Netherlands, p179.

Campbell, E. Bate, G. (1992) Groundwater in the Alexandria dune field and its potential

influence on the adjacent surf-zone, Water SA, 17, pp 155-160.

Capone, D. & Slater, J. (1990) Interannual patterns of water table height and groundwater

derived nitrate in nearshore sediments, Biogeochemistry 10, pp.277-288.

II

Capone, D.G. & Bautista, M.F. (1985) A groundwater source of nitrate in near shore

marine sediments, Nature, vol.3.3, pp. 214-217

Chapman, VJ (1976) Mangrove vegetation. J Cramer Verlaa, Vaduz, Liechtenstein.

Chebotarev, I,I.(1955) Metamorphism of natural waters in the crust of weathering,

Geochim, Cosmochim, Acta 8, pp.22-212

Cherry, J.A., Gillham,R.W. (1978) Field evidence of denitrification in shallow

groundwater flow systems, Proceedings of the 13th Canadian Symposium on Water

Pollution Resources, Mcmaster University, Hamilton, Ontario.

Dominico, P.A. (1972) Concepts and Models in Groundwater Hydrology, McGraw-Hill,

New York

Drever, JI. (1988) The geochemistry of natural waters. Second edition. Prentice Hall, New

Jersey.

Driscoll, F.G. (1986) Groundwater and Wells, Johnson Division, Minnesota.

Faye, S.; Maloszewski, P.; Stichler, W. & Trimborn, P. (2004) Groundwater Salinization

in the Saloum (Senegal) delta aquifer: minor elements &isotopic indicators, Science of

The Total Environment, Vol.343, pp243-259

III

Flores, G (1973) The cretaceous and tertiary sedimentary basins of the African Coasts,

Association of African Geological surveys, Paris.

Freeze, RA & Cherry, JA (1979) Groundwater, Prentice-Hall Inc.

Ghazvinian, Amir. (2004) Impacts of deforestation in Mozambique, Seattle: Village reach

Gundry, J.M., Drennan, M. & Bejark, P. (1981) Salinity control in relation to pre-

abscission propagule development in Avicennia marina, Electronic Microscopy Society of

south Africa – Proceedings11, pp. 155-156

Ridd, P.V. (1996) Mangrove Muds – A sink and a source for trace metals, Marine

Pollution Bulletin Vol.17, pp.246-250

Hoguane, A.M., Hill., A.E., Simpson. J.H & Bowers. D.G. (1999) Diurnal & tidal

variation of temperature and salinity in the Ponta Rosa Mangrove Swamp, Mozambique,

Estuarine, Coastal and Shelf Science, Vol. 4, pp 251-264.

Hoguane, AM., Marine Science and Oceanography in Mozambique: Overview of

Mozambique, http://www.aaas.org./international/africa/moz/html

http://villagereach.org/Impacts%20of%20deforestation%20in%20mozanbique.pdf

IV

Hughes, C.E., Binning, P.H. & Willgoose, G.R. (1998) Characterisation of the Hydrology

of an estuarine wetland, Journal of Hydrology, Vol. 211, pp. 34 – 49.

IWACO (1986), Study of groundwater to supply Maputo.

Johannes, R.E. (1980) The ecological significance of the submarine groundwater

discharge of groundwater

Lanyon, J.A., Eliot, I. & Clark,D.J. Groundwater level variation during semi-diurnal

spring tidal cycles on a sandy beach in Australia, Journal of Marine Freshwater Resources

33, pp.377-400

Li, L., Barry, D.A., Cunning, C., Stagnitti, F. & Parlange, J.Y A two-dimensional

analytical solution of groundwater responses to tidal loading in an estuary and ocean,

Advances in Water Resources 23, Vol.8, pp. 825-833.

Lloyd, J.W. & Heathcote, J.A. (1985) Natural inorganic hydrochemistry in relation to

groundwater, an introduction. Clarence Press, Oxford.

Matimula, J.(2003) General characteristic of Mozambique, In children environmental

health country files: Mozambique. http://www.afro.who.int/eph/ceh/country-

profile/mozambique.pdf

V

Mendes, M. (1998) Best practice of environmental information systems(EIS: the case of

Mozambique.http://easd.org.za/Eis/repts/Mozambique/mozeis1htm

Midgley, J.J. 7 Scott, D.F. (1994) The use of stable isotopes of water in hydrological

studies in the Jonkershoek Valley, Water SA, 20, pp. 151-154.

Reilly, T.E., Goodman, A.S. (1985) Quantitative analysis of saltwater-freshwater

relationships in groundwater systems, Journal of Hydrology, Vol. 80, pp. 125-160.

Ridd, P.V. (1996) Flow through animal burrows in mangrove creeks, Estuarine, Coastal &

Shelf Science, Vol.43, pp.617-625.

Riedel, J., Gerhardr, F., Sanders, G. & Osman, R. (1997) nBiogeochemical control on the

flux of trace metals from estuarine sediments, Estuarine, Coastal & Shelf Science 44,

pp.23-38.

Rust, I.C. (1987) Coastal Aquifer characteristics : St. Francis Bay and Plettenberg Bay, In:

Proceedings of the 1987 Hydrological Science Symposium, Vol I, Hydrological Research

Institute, Rhodes University.

Seliger, H., Boggs, J. & Biggley, W.H. (1985) Catastrophic anoxia in the Chesapeake Bay

in 1984, Science 228, pp.70-73.

VI

Serfes, M.E.(1991) Determining the mean hydraulic gradient of groundwater affected by

tidal fluctuation, Groundwater 29, Vol.4, pp.549-555.

Simons, G.M. (1992) Importance of submarine groundwater discharge and seawater

cycling to material fluxes across sediment / water interfaces in marine environments,

Marine Ecology Progress Series, vol 84, pp. 173 – 184.

Smith, T.J. (1988) Differential distribution sub-species of the mangrove ceriops tagal:

competitive interactions along a salinity gradient, Aquatic Botany 32, pp. 79-89.

Staver, K.W. & Brinsfield, R.B. (1996) Seepage og groundwater nitrate from a riparian

agrosystem into the Wye River Estuary, Estuaries 19, Vol.2B, pp.359-370.

Todd, D.K. & Mays, L.W, Groundwater Hydrology, Third Edition, John Wiley & Sons,

Inc.

Tomlinson, CB (1986) The Botany of Mangroves. Cambridge Tropical Biological

Science, Cambridge University Press, Cambridge, New York, USA

Valiela, I., Costa, J., Forman, K., Teal, J.M., Howes, B. & Audrey, D. (1990) Transport of

water-borne nutrients from watershed and their effects on coastal waters,

Biogeochemistry 10, pp. 177-198.

VII

Vance, D.J. , Haywood, M.D.E. & Staples, D.J. (1990) Use of a mangrove estuary as a

nursery area by post larval & juvenile banana pawns in northern Australia, Estuarine,

Coastal & Shelf Science 31, pp.689 – 701.

Weaver, J.M.C. (1992) Groundwater sampling, a comprehensive guide for sampling

methods, Report no. TT54/92, Water Research Commission.

Weaver, J.M.C., Talma, A.S. & Cavé, L.C. (1999) Geochemistry and isotopes for

resources evaluation in the Fractured Rock aquifers of the Table Mountain Group, Report

no. 481/1/99, Water Research Commission.

Wolanski, E. (1992) Hydrodynamics of mangrove swamps and their coastal waters,

Hydrobiology 247, pp.141-161

Wright, A. (1991) The importance of establishing a meaningful water quality monitoring

programme with any any coastal groundwater supply scheme, Proceedings of the Biennial

Groundwater Convention, Groundwater Division of the Geological Society of South

Africa, Midrand

Zektzer, I.S., Ivanov, V.A. & Meskheteli, A.V. (1973) The problem of direct discharge to

the seas, Journal of Hydrology, Vol. 20, pp. 1 – 36

VIII

List of Abbreviations

C2C : Catchment -2- Coast

DNA : National Water Directorate, Mozambique

CSIR : Council for Scientific and Industrial Research

DWAF : Department of Water Affairs and Forestry

WHO : World Health Organisation

IWACO : Consultants for Water and Environment

IX

APPENDIX: Methods

Water analysis 1. pH measurement

Principle

The basic principle of electrometric pH measurement is determination of the activity of the

hydrogen ions by potentiometric measurement using a standard hydrogen electrode. The hydrogen

electrode consists of a platinum electrode across which hydrogen gas is bubbled at a pressure of 101

kPa. Because of difficulty in its use and the potential for poisoning the hydrogen electrode, the glass

electrode is commonly used. The electromotive force (emf) produced in the glass electrode system

varies linearly with pH. This linear relationship is described by plotting the measured emf against

the pH of different buffers. Sample pH is determined by extrapolation. This method is applicable to

the determination of pH in all water samples.

Reagents

a) Saturated potassium chloride solution.

b) Reference buffer solutions.

(i) pH 4, 00 @ 25° C - potassium hydrogen phthalate

Dry 15g KHC8H404 at 105° C for two hours and cool in a dessicator. Dissolve 10,12g in

water and dilute to 1000mL.

(ii) pH 6, 86 @ 25° C - phosphate reference buffer solutions

Dry KHZP04 and Na2HP04 at 105°C for two hours and cool in a dessicator. Dissolve 3,39g

KHzP04 and 3,53g Na2HP04 in water and dilute to 1000mL.

pH 10, 01 @ 25° C - sodium bicarbonate - sodium carbonate reference buffer solution. Dry

Na2C03 at 105° C for two hours and cool in a dessicator. Dry Na2C03 and 2,092g NaHC03 in

water and dilute to 1000mL.

X

Equipment A Radiometer PHM82 STANDARD pH METER and GK 2401C combined electrode is used

together with a stirrer.

Interference

The glass electrode is relatively free from interference from colour, turbidity, colloidal matter,

oxidants, reductants or high salinity, except for a sodium error at a pH > 10. Samples with low

buffering capacity are susceptible to pH changes.

Calibration

Use two buffers to calibrate the meter, usually pH 7,00 or 6,86 and pH 4.01.

Select the pH mode (released push button). Select a pH buffer 7 or 6.86 and dip electrode in the

buffer and switch on stirrer. Leave stirrer on until pH measurements are completed. After stable

reading is obtained, turn the BUFFER knob to make the PHM82 indicate the pH of the buffer

and then rinse the electrode with deionised water.

Select pH buffer 4.01 and dip electrode in the buffer and allow to stabilise, while stirring.

Adjust by turning the SENS.adjuster on the meter until the display indicates 4,00. Rinse

electrode.

Analytical procedure

Rinse electrode and immerse in the sample to be tested.

Read the pH of the sample while stirring. Record the results and after use, store the electrode in

a pH 4 buffer solution.

Calculation of results

The measured pH is read directly from the instrument display. Report the results to the nearest

0, 1 pH unit.

XI

2. Electrical Conductivity

Principle

Conductivity is a numerical expression of the ability of an aqueous solution to carry an

electric current. The physical measurement made in a laboratory determination is

usually of resistance. The resistance of a conductor is inversely proportional to its

cross-sectional area and directly proportional to its length. The magnitude of the

resistance measured in an aqueous solution therefore depends on the characteristics of

the conductivity cell used, and is not meaningful without knowledge of these

characteristics.

The term conductivity" is preferred and is usually reported in microhms per centimeter. -

The SI units use the Siemens (the reciprocal of the ohm) and conductivity is reported as

milli-siemens per meter (mS/m). 1 mS/m = 10 micro ohms/cm. To report results in SI

units, divide phos/cm by 10.

Reagents

Dried potassium chloride(KCI) at 105- C for two hours and cooled in a dessicator. To

prepared a solution of a particular conductivity, select the corresponding mass of KCI

from the table below. Weigh out the exact mass and transfer quantitatively to 1 litre

volumetric flask. Dissolve the appropriate mass of KCI in approximately 500m1 reagent

grade water and dilute to 1000mL.

M KCI mass KCI Conductivity mS/m @ 25° C

0,005 0,3728 71,8

0,01 0,7456 141,2 0,02 1,4912 276,5

0,05 3,7280 666,7

0.1 7,4560 1289,0 Table 2A: Potassium chloride solutions KCI

XII

Equipment

Conductivity meter with platinum electrode cell and a thermometer

Interference

Electrolytic conductivity (as opposed to metallic conductivity) increased with temperature at a

rate of approximately 1,9% °C. Significant errors can result from inaccurate temperature

measurement.

Analytical Procedure

Rinse cell with one or more portions of sample. Adjust temperature. Measure sample

conductivity.

Calculation of results

Direct reading @ 25° C from conductivity meter.

3. Alkalinity

Principle The alkalinity of water is defined as the capacity that some substances have to take up protons In

other words, to react with an equivalent quantity of a strong acid. Examples of such substances

are hydroxide ions and anions of weak acids, eg. bicarbonate, carbonate, phosphate and silicate.

The equivalent quantity of strong acid required to neutralise these ions is equal to the total

alkalinity. The alkalinity is calculated by titrating a known volume of sample to a predetermined

endpoint, using an auto-tritrator.

Reagents

Standard Hydrochloric acid 0,01 N.

Disodium carbonateNa2CO was dried at 105°C for two hours and cooled in a dessicator. 2,65g

Na2CO was weighed out and dissolved in 1000mL C02 free water. This stock solution has an

XIII

alkalinity of 2500mg/L as CaC03. The stock solution was diluted further to make working

solutions.

Equipment A pH meter that can be read to 0,05p1-I units equipped with a combined pH electrode.

(Radiometer PHM 82), a stirrer and a Radiometer Auto burette ABU 80 was used in

conjunction with Radiometer Autotitrator unit TTT 80)-.

Interferences

Soaps, oily matter, suspended solids or precipitates may coat the glass electrode and cause

sluggish response. Allow additional time between additions to let electrode come to equilibrium

or clean the electrodes occasionally. Do not filter, dilute, concentrate or alter sample.

Analytical Procedure

• Rinse Autoburette with acid used for titration. Calibrate pH meter using 6.86 and 4.0 pH

buffers.

• Potentiometric titration to pre-selected pH

• An endpoint pH of 4,5 was used.

Transferred 20mL sample with a volumetric pipette to a beaker and placed it on the stirrer.

Placed the electrode and tip of the burette in the sample. Titrated the solution with the standard

0,010N HCI to the endpoint pH of 4,5. Note the titration reading down.

Calculation of results

Total Alkalinity as CaC03 mg/L = mL 0,01N HCI x 0,01 x 1000 x 50 mL sample volume

XIV

4. Major cations : Calcium, Magnesium, Potassium and Sodium

Principle

Atomic absorption spectrometry resembles emission flame photometry in that a sample is

aspirated into a flame and atomised. The major difference is that in flame photometry the amount

of light emitted is measured. Whereas in atomic absorption spectrometry a light beam is directed

through the flame, into a monochromator and onto a detector that measures the amount of light

absorbed by the atomised element in the flame. For some metals, atomic absorption exhibits

superior sensitivity over flame emission. Because each metal has its own characteristic absorption

wavelength, a source lamp composed of that element is used, this makes the method relatively

free from spectral or radiation interferences. The amount of energy at the characteristic

wavelength absorbed in the flame is proportional to the concentration of the element in the

sample over a limited concentration range.

Reagents

a) Cesium -lanthanum interference suppressant solution

Added 2L deionised water and 1L hydrochloric acid to a 5L volumetric flask. Added

60g La2O and CsCI (both AR grade), and shook to dissolve. Allowed solution to cool

down to room temperature and made it up to the mark. This solution contains 5,5g/L Cs

and 10,2g/L La.

b) Standard stock solutions

Use analytical grade standard solutions (1000mg/L) for the preparation of working standard

solutions.

i) Preparation of stock mixed standard solution (100mg/L Ca. 100mg/L Na, 30mg/L K

and 30mg/L Mg) Add 200mL each of 1000mg/L Ca and Na standard, and 60mL each of 1000mg/L K and Mg standard into a 2000mL

volumetric flask. Add 200mL of the La/Cs solution and made it up to the mark.

ii) Preparation of working standards

Add to 100mL volumetric flasks the following volumes of stock mixed standard:

XV

Table 4A: Preparation of major cation working standards

Volume Added mL La/Cs added Working standard concentration

Na Ca K Mg 2,5 10 2,5 2,5 0,75 0,75 5 10 5 5 1,5 1,510 10 10 10 3 320 10 20 20 6 630 10 30 30 9 940 10 40 40 12 1250 10 50 50 15 1560 10 60 60 18 1870 10 70 70 21 2180 10 80 80 24 24100 - 100 100 30 30

Equipment

A flame atomic absorption spectrometer consisting essentially of the following components was

used :

(i) A stable light source, emitting the sharp resonance line of the element to be

determined.

(ii) A flame system, into which the sample solution may be aspirated at a steady rate, and which

is of sufficient temperature to produce an atomic vapour of the required species from the

compounds present in the solution.

(iii) A monochromator to isolate the resonance line and focus it upon a photomultiplier.

(iv) A photomultiplier to detect the intensity of light energy falling upon it, which is followed by

facilities for simplification and readout.

(v) A chart recorder for recording the output.

Interference

In the field of water analysis, the most commonly encountered interferences are chemical and

ionization interferences. Other types, less commonly encountered; include non-atomic

XVI

absorption and matrix interferences. Chemical interferences occur when the element being

determined combines chemically with another reactive component in the sample. The resulting

compound influences the atomization process in the flame, thus altering the number of free

atoms available to absorb light. Chemical interferences can be controlled by either:

- Addition of a releasing agent: A releasing agent is a chemical which reacts preferentially with

either the element being determined or with the interfering component. A common example is

the addition of a solution of a lanthanum salt to calcium solutions to overcome the depressing

effect of phosphate on the calcium signal.

- Use a higher temperature flame: this can overcome many interferences because more energy is

available to break down compounds which would be stable in cooler flames For example, the

interference of phosphate on calcium observed in the air-acetylene flame is eliminated in the

nitrous oxide-acetylene flame.

Ionization interferences occur when the flame temperature is high enough to ionize- a

significant fraction on the element being determined. This lowers the number of atoms which

can absorb radiation and reduces the analytical signal. Analytical errors can occur when the

sample and standards exhibit different degrees of ionization. The simplest way to control

ionization interferences is to add a large excess of an easily ionisable cation, such as cesium, to

both samples and standards. The electrons provided by the more easily ionized element

combine with the ions of the element being determined, and increase the number of atoms

which can absorb radiation. A common example is the addition of a solution of ~a cesium salt

to barium solutions to overcome the ionization of this metal in the nitrous oxide-acetylene

flame. The use of lower temperature flame can also control ionization effects. For example,

sodium is partially ionized in the air-acetylene flame, but ionization is negligible in the air-

propane flame.

(c) Non-atomic absorption results from the absorption of radiation from the hollow cathode

lamp by materials in the flame other than the element of interest. It may be due to either:

(i) Molecular absorption, which is caused by the presence of molecular species which absorb

light at the same wavelength as that of the element being determined.

(ii) Light scattering by particles in the flame, which occurs when solutions containing high

amounts of dissolved solids are aspired into the burner?

XVII

Non-atomic absorption maybe controlled by the use of a `continuum source corrector', which

may be either a hydrogen-filled hollow cathode lamp or a deuterium arc lamp. Matrix

interferences occur when the sample matrix is so complex that viscosity, surface tension

and components cannot be accurately matched with standard, resulting in the uptake

rate, nebulization efficiency and atomization processes in the flame being affected.

causing erroneous results. Matrix interferences may be controlled by use of standard

addition techniques or by means of MIBK extractions with APDC, which are particularly

useful with seawater samples, for example.

Analytical Procedure

i) Pretreatment

Major cations in natural waters are usually in proportion to the total dissolved salts of

that water. In other words, by determining the electrical conductivity of a sample, an

appropriate dilution can be made so that the sample falls within the range of the working

standards; the following table is used:

Conductivity

mS/m

Volumetric Flask

size (mL)

Pipette

size (mL)

Volume La/Cs

(mL

Dilution

factor 0 - 70 25 20 2,5 1,25

70 - 90 25 15 215 1,66690 - 110 50 25 5 2110 - 135 25 10 2,5 2,5135 - 180 50 15 5 3,333180 - 270 50 10 5 5270 - 360 100 15 10 6,666360 - 450 100 10 10 10450 - 900 100:50 10:25 5 20900 - 1125 100:25 10:10 2,5 251125 - 1500 100:50 1015 5 33,3331500 - 2250 100:50 10:10 5 502250 - 3000 100:100 10:15 10 66.663000 - 4500 100:100 10:10 10 100

4500 + 100:100 15:5 10 133,33

Table 4B: Dilution table for cation analysis

XVIII

A list of recommended wavelengths and flame gases for each metal determined is shown

below.

Element Wavelength (nm) Flame gases

Calcium 422,7 Air-acetylene or nitrous oxide-

Magnesium 285,2 Air-acetylene or nitrous oxide-

Potassium 766,5 Air-acetylene

Sodium 589,5 Air-acetylene

Table 4C: Recommended wavelength and flame gases

Because of the difference between makes and models of atomic absorption spectrometers,

it is not possible to formulate instructions applicable to every instrument. The

manufacturer's instructions for each particular instrument should be followed. In general;

however, the following procedure is employed (air-acetylene flame):

(a)Install the hollow cathode lamp for the metal being measured and roughly set the

desired wavelength (see Table 1).

(b) Set the slit width and lamp current at the values suggested in the manufacturer's instructions

and allow the instrument to warm up until the energy source becomes stable (10 - 20 minutes).

(c) Readjust the current as necessary after warm up.

(d) Optimize wavelength by adjusting wavelength dial until optimum energy gain is obtained.

(e) Align lamp in accordance with manufacturer's instructions.

(f) Install air-acetylene burner and adjust burner head position. Turn on air and adjust flow rate

to that specified by manufacturer to give maximum sensitivity for the metal being measured.

(g)Turn on acetylene, adjust flow rate to value specified, and ignite flame.

(h) Aspirate a standard solution of the desired metal and adjust aspiration rate of the nebulizer

(if variable) to obtain maximum sensitivity. Aspirate a standard solution near the middle of the

linear working range and adjust the burner vertically and horizontally to obtain maximum

response.

(I)The instrument is now ready for operation.

On completion of the analysis, extinguish flame by turning off first acetylene and then air.

XIX

However, for determination involving the use of a nitrous oxide-acetylene flame, proceed as in

(a) to (c) above and then continue as follows:

(a) Install nitrous oxide-acetylene burner and adjust burner head position.

(b) Turn on acetylene (without igniting flame) and adjust flow rate to specified value, then turn

off acetylene.

(c) With both, air and nitrous oxide supplies turned on, set the T-Junction valve to nitrous oxide

and adjust flow rate to specified value.

(d) Turn the switching value to the air position and check that flow rate is the same. Turn on

acetylene and ignite to a bright yellow flame. With a rapid motion, turn switching value to

nitrous oxide.

(e) The flame should now have a red cone above the burner. If not, adjust the fuel flow to

obtain a red cone. After ignition, allow the burner to come to thermal equilibrium before

commencing analysis.

(f) Aspirate a standard solution of the desired metal and adjust aspiration rate of the nebulizer

(if variable) to obtain maximum sensitivity. Aspirate a standard solution near the middle of the

linear working range and adjust the burner vertically and horizontally to obtain maximum

response.

(g)The instrument is now ready for operation.

On completion of the analysis, extinguish flame by first turning the switching valve from

nitrous oxide to air and then turning off the acetylene. This procedure eliminates the danger of

flashback occurring on direct ignition or shut down of nitrous oxide and acetylene.

5. Dissolved Organic Carbon (DOC)

Principle

The method utilises an automated persulphate -ultraviolet oxidation method using

calorimetric detection of CO2 generated. Organic carbon is oxidised to CO2 by

XX

persulphate in the presence of ultra violet light. The C02 produced diffuses through a

gas permeable membrane and is measured by the decrease in absorbency of a

phenophthalein solution. Inorganic carbon is removed by acidifying the sample to pH 2

or less to convert the inorganic species to CO2 and then purging the sample with a

purified gas (N2) to strip off the CO2.

Reagents

(a) Stock Organic Carbon solution

Dissolve 2.1254g potassium biphthalate (dried at 105°C for 24 hours) in 500mL and

make up to 1000mL with water. Dilute 40mL of stock solution to 200mL = 200mg/L

as C. Prepare working standards in 200mL volumetric flasks as follows:

Table 5A: Preparation of DOC working standards

mL 200 mg/L in 200mL Flask 5,0 10,0 15,0 20,0

Standard conc. mg/L as C 5,0 10,0 15,0 20,0

(b) Potassium persulphate solution

Dissolve 24g potassium persulphate in 800mL water and dilute to 2000mL.

(c) Phenolphthalein 1 %

Dissolve 1,Og of phenolphthalein in 80mL methanol and dilute to 100mL with methanol.

(d) Sodium Carbonate 0,1 M

Dissolve 10,6g sodium carbonate in 800mL deionised water and make up to 1000mL.

XXI

(e) Sodium bicarbonate 0,1 M

Dissolve 8,4g sodium bicarbonate in 800mL deionised water and make up to 1000mL.

(f) Carbonate-Bicarbonate Buffer

Mix 150mL of the 0.1M sodium solution with 300mL of the 0.1 M sodium bicarbonate

solution. Use this solution for the preparation of the colour reagent.

(g) Phenolphthalein colour reagent

Pipette 1,0mL of the 1 % Phenolphthalein solution into a 1000mL volumetric flask. Add

800mL deionised water and 15,0mL of the carbonate-bicarbonate buffer mixture and

0,5mL of Brij 35. Dilute to 1000mL with deionised water.

(h) Hydroxylamine ammonium chloride

Dissolve 70g of hydroxylamine ammonium chloride in 800mL deionised water, add

28mL of conc. sulphuric acid, mix well and make up to 1000mL with deionised water.

Equipment

Automated system consisting of the following:

Recorder, Calorimeter, Mixing manifold with appropriate coils, proportioning pump and

an automatic sampler.

Interferences

Persulphate oxidation of organic molecules is slowed in the presence of chloride by the

preferential oxidation of chloride; at a concentration of 0,1% chloride, oxidation of

organic matter may be inhibited completely. To remove this interference hydroxylamine

ammonium chloride is added

Analytical Procedure

Set up manifold and complete system as shown in figure 7.

Calculation of results

XXII

Prepare standard curves by plotting peak heights of standards processed through the manifold

against DOC concentrations in standards. Compute sample DOC concentration by comparing

sample peak height with standard curve.

6. Chloride and Sulphate

Principle of the chloride method

Thiocyanate ion is liberated from mercuric thiocyanate by the formation of soluble mercuric

chloride. In the presence of ferric ion, free thiocyanate ion forms a highly coloured ferric

thiocyanate, of which the intensity is proportional to the chloride concentration. There are no

significant interference, however turbid samples needs to be filtered.

Principle of the sulphate method

The method is based on the formation of low solubility barium sulphate(BaSO4) suspension in a

gelatine medium. (Ksp of BaSO4 = 9.2*10-11). The degree of suspension of the BaSO4 is

dependent on the reaction condition and it is necessary to ensure that the conditions that are

required wit regard to the reagents must be strictly followed. The addition of hydrochloric acid

prevents the formation of precipitates of sulphite, carbonate and phosphate with barium.

Reagents

A mixed stock standard containing 2000mg/L chloride and 1000mg/L sulphate was prepared.

The stock standard was prepared by dissolving 6.5928g sodium chloride (NaCl) and 2.9574g

sodium sulphate (NaSO4) in deionised water and made it up to 2000ml. From this stock solution,

the working standards were prepared in 200ml volumetric flasks as follows:

Table 6A: Preparation of Cl and SO4 working standards

mL of

2000mg/L Cl

and1000mg/L

2.5 5 10 15 20 30 40 50

XXIII

SO4 in a

200ml flaks

Standard

conc. of

Cl(mg/L)

25 50 100 150 200 300 400 500

Standard

conc. of

SO4(mg/L)

12.5 25 50 75 100 150 200 250

a) Stock mercuric thiocyanate solution for chloride analysis

Dissolved 4.17g mercuric thiocyanate Hg(SCN)2 in about 500mL methanol, dilute to 1000ml with

methanol. Mix and filter.

b) Stock ferric nitrate solution for chloride analysis

Dissolved 202g ferric nitrate ( Fe(NO3)3.9H2O) in 500ml deionised water. Carefully added 21ml

conc. nitric acid. Dilute to 1000ml with deionised water and stored solution in an amber bottle.

c) Colour reagent for chloride analysis

Add 150ml stock ferric thiocyanate to 150ml stock ferric nitrate solution, mix and dilute to

1000ml with deionised water. Stored solution in an amber bottle.

d) Buffer solution for sulphate analysis

Dissolve 40g EDTA, 7g ammonium chloride and 57mL ammonium hydroxide in 800mL water

and make up to 1000mL.

e) Barium chloride - Gelatin solution for sulphate analysis

Boil 200mL water in a 1000mL beaker, add 2g gelatin and stir until dissolved. Add 20g BaCl2

and stir to dissolve. Add 10mL concentrated hydrochloric acid and make up to 1000mL

XXIV

Equipment

The autoanalyser system used consists of the following:

A recorder, colorimeter, mixing manifold with appropriate coils, proportioning pump and an

automatic sampler.

Analytical procedure

Major anions (sulphates and chlorides) in natural waters are usually in proportion to the total

dissolved salts of the water. So by determining the conductivity of a sample, an appropriate

dilution can be used so that the sample falls within the range of the working standards. The

following table was used:

Table 6B: Dilution table for Cl and SO4 analysis

Conductivity

(mS/m)

Volumetric flask

size(mL)

Pipette

size(mL)

Dilution

factor

0-170 No dilution necessary 0 0

170-280 50 25 2

280-500 50 15 3.33

500-700 50 10 5

700-1200 100 10 10

1200-2000 200 10 20

2000-4000 500 15 33.33

4000- 500 10 50

Switched on the chart recorder and colorimeter. Allowed the system to warm up until a stable

baseline is obtained.

Calculation of results

XXV

Prepared a standard curve by plotting peak heights of standards processed through the manifold

against the chloride and sulphate concentrations by comparing sample peak height with standard

curve.

7. Orthophosphate

Principle of the method Ammonium molybdate and potassium antimonyl tartrate react with orthophosphate in an acid

medium to form n antimony-phosphomolybdate complex, which on reduction with ascorbic acid,

yields an intense blue colour suitable for photomeric determination.

Reagents:

A mixed stock solution for ammonia, nitrate and phosphate was prepared with a concentration of

1000mg/L. This solution was prepared by dissolving 4.393g anhydrous potassium

dihydrogenphosphate (KH2PO4), 3.819g anhydrous ammoniumchloride (NH4Cl) and 7.218g

potassium nitrate(KNO3) in deionised water and made up to 1000ml, after it has been dried at

1050C. This solution has a concentration of 1000mg/L each for NH4-N, NO3-N and PO4-P.

Diluted 20ml of this 1000mg/L solution to 1000ml to obtain a solution of 20g/L each of NH4-N,

NO3-N and PO4-N. The working standards were prepared from this solution in 200ml volumetric

flasks as follows:

Table 7A: Preparation of Orthophosphate working standards

ml of 20mg/L in 200ml flask 5 10 15 25 50 75 100

Standard conc. In mg/L 0.5 1 1.5 2.5 5 7.5 10

a) Potassium antimonyl tartrate solution

Dissolved 0.3g K(SbO) C2H4O6 . ½ H2O in 50ml deionised water and diluted it to 100ml. The

solution needs to be stored at 4oC in an amber bottle with a glass stopper.

XXVI

b) Ammonium molybdate solution

Dissolved 4g (NH4)6MoO24.4H2O in 100ml water and stored it at 4oC in a plastic bottle.

c) Ascorbic acid solution

Dissolved 2g ascorbic acid in 100ml deionised water and stored the solution at 4oC. This solution

is only stable for a month.

d) Dilute sulphuric acid

Slowly added 10ml of conc. H2SO4 to 600ml deionised water and diluted it to 1000ml when

cooled.

e) Combined reagent Mixed the above reagents in the following proportions for 100ml of combined reagents:

50ml dilute 5N sulphuric acid (H2SO4), 5ml potassium antimonyl tartrate, 15ml ammonium

molybdate and 30ml ascorbic acid solution. Mix after the addition of each reagents. This solution

is only stable for one day.

Equipment

An autoanalyser system consisting of the following was used:

Recorder, colorimeter, mixing manifold with appropriate coils, proportioning pump and an

automatic sampler.

Analytical procedure

Set up manifold and complete system as shown in figure c

Switched on chart recorder and colorimeter and allowed system to warm up or until a stable

baseline is obtained.

Calculation of results Prepared standard curves by plotting peak heights of standard processed through the manifold

against PO4-P concentrations in standards. Compute sample P04-P concentration by comparing

sample peak height with standard curve.

XXVII

8. Ammonia

Principle of the method

Alkaline salicylate and hypochlorite react with ammonia to form a green-blue colour, the

absorption of which is proportional to the ammonia concentration. The colour is intensified by the

addition of sodium nitro prusside.

Interference

In alkaline solution calcium and magnesium will interfere by forming a precipitate. Sodium

citrate buffer prevents this interference.

Standards

Stock solution

Dry NH4Cl at 105° C for two hours and a cool in dessicator. Dissolved 3,819g NH4C1 in water

and dilute to 1000mL. Dilute 20mL of stock solution to 1000mL to give 20mg/L.

Working standards

Prepare working standards in 20mL volumetric flasks as follows.

mL 20mg/L in 200mL flask

5

10

15

25

50

Standard concentration mg/L

0,5

1,0

1,5

2,5

5,0

Table 8A: Preparation of ammonia working standards

Reagents

a) Sodium salicylate solution

Dissolve 90g of sodium salicylate in 800mL water and make up to 1000mL.

b) Sodium nitroprusside solution

Dissolve 0,5g of sodium nitroprusside in 800mL water and make up to 1000mL.

XXVIII

c) Sodium hypochlorite solution

Dilute 75mL of commercial bleach (Javell) to 250mL with water.

d) Buffer solution

Sodium Citrate 300g, EDTA 10g, Sodium hydroxide 5g

Dissolve these chemicals in deionised water and make up to 1000mL with water.

Equipment

Auto Analyser system consisting of the following:

Recorder, Colorimeter, Mixing manifold with appropriate coils, Proportioning pump and an

automatic sampler.

Analytical Procedure

Switch on pump, colorimeter, chart recorder and allow to stabilize so as to obtain a stable

baseline with all reagents, feeding water through the sample line.

Calculation of results

Prepare standard curves by plotting peak heights of standards processed through the manifold

against NH4 - N concentrations in standards. Compute sample NH4 - N concentration by

comparing sample peak height with standard curve.

9. Nitrate

Principle of the method

N03 - is reduced almost quantitatively to N02 - in the presence of cadmium. This method uses

commercially available Cd granules treated with CuS04 and packed in a column. The NO2 - thus

produced is determined by diazotising with sulphanilamide and coupling with N - (1 - naphthyl) -

ethlenediaminehydrochloride to form a highly coloured azo dye that is measured colorimetrically.

XXIX

A correction may be made for any NO2- in the sample by analysing without the reduction column

step.

Reagents

a) Stock solution

Dissolve 7,218g KN03 (dried at 105° C for two hours) in distilled water and make up to

1000mL.This solution contains 1000mg/L NO3-N.

Working standards

Prepare working standards in 200 mL volumetric flasks as follows:

mL 20 mg/I in 200mL Flask

5

10

15

25

50

75

100

Standard Concentration mg/L 0.5 1.0 1.5 2.5 5.0 7.5 10.0

Table 9A: Preparation of nitrate working standards

b) Copper sulphate solution

Dissolve 20g CuS04. 5H2O in 500mL and dilute to 1000mL with water.

c) Ammonium chloride solution

Dissolve 85g NH4C1 in water and dilute to 1000mL. Add to 0,5mL Brij 35.

d) Colour Reagent

Add to 800mL water while stirring, 1000mL conc. phosphoric acid, 40g sulphanilamide, and 2g

N-(1naphthyl) - ethylenediamine hydrochloride. Stir until dissolved and dilute to 1000mL. Store

in brown bottle.

XXX

e) Copper-Cadmium granules

Wash 25g new or used 40-60 mesh Cd granules with 6N HCI and rinse with water. Swirl Cd with

100mL 2% CuS04 solution for 5 min or until blue colour partially fades. Decant and repeat with

fresh CuS04 until a brown colloidal precipitate begins to develop. Gently flush with water to

remove all precipitate Cu.

Equipment

Auto Analyser system consisting of the following:

Recorder, Colorimeter, Mixing manifold with appropriate coils and Cadmium reduction column,

proportioning pump and an automatic sampler.

Analytical Procedure

Preparation of cadmium reduction column

Insert a glass wool plug into the bottom of the reduction column and fill with distilled water. Add

sufficient Cu-Cd granules to produce a column 18,5cm in length. Switch on pump, colorimeter,

chart recorder and allow to stabilize so as to obtain a stable baseline with all reagents, feeding

water through the sample line.

Calculation of results

Prepare standard curves by plotting peak heights of standard processed through the manifold

against N03-N concentrations in standards. Compute sample N03-N concentration by comparing

sample peak height with standard curve.